research-map-digital-design.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![Map of Digital Design in Architecture](images/map-digital-design-in-architecture.png)\n",
    "\n",
    "This is a map of **Digital Design in Architecture**. It is created with data from 12,000 articles published between 1975 and 2019.\n",
    "\n",
    "You can download a [high-res PDF version here](images/map-digital-design-in-architecture.pdf).\n",
    "\n",
    "You can use this map to learn about the various concepts that build the field of digital design in architecture and the main authors defining them.\n",
    "\n",
    "You can also check what is not there yet.\n",
    "\n",
    "The code i used to generate it is below."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# IDEA\n",
    "\n",
    "I recently found this interesting [visualization of 20th Century Philosophy by Maximilian Noichl](https://homepage.univie.ac.at/noichlm94/posts/structure-of-recent-philosophy-iii/).\n",
    "The chart is drawn by sampling and clustering some 50,000 peer-reviewed journal article on philosophy. \n",
    "It was amazing how I was able to discover relationships and authors in a discipline, which I am not familiar with, through the data clustering that it presented!\n",
    "\n",
    "So I wanted to do a similar thing for my field of research for two reasons:\n",
    "- I feel at home working in a data-driven workflow. So it felt natural to try and help myself structure the sourcing process for the literature review for my PhD thesis using a data-driven approach as well. \n",
    "- I was curious what discussion such a visualization would prompt in our computational design community. \n",
    "\n",
    "Maximilian has generously shared his code and when I emailed him, he advised me on what to pay attention to when creating my own map. Big thanks to Max!\n",
    "I used Max's code as a base to built this map. My changes were in the cluster tree chart and in the text analysis. I also documented the process on how to get the data needed for such a map.\n",
    "\n",
    "The idea of this Jupyter notebook is that you can use it to produce your own map for your own field of interest. It is build step by step but is meant to be used in a loop where you would see the result in one step and go back few steps to change the parameters (for example the size of the research clusters - Max used 440, in my case this number was too big because I had less articles) and run again."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# INIT\n",
    "Let's initialize the environment and load the needed libraries. \n",
    "Check the [SETUP.md](SETUP.md) file for instructions on how to set up your development environment so that you can run this Jupyter notebook."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# This cell loads the needed Python libraries\n",
    "import metaknowledge as mk\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "from random import randint\n",
    "import datetime\n",
    "import scipy as scipy\n",
    "\n",
    "import csv\n",
    "import textwrap\n",
    "\n",
    "%matplotlib inline\n",
    "\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "#For Tables:\n",
    "from IPython.display import display\n",
    "from IPython.display import Latex\n",
    "pd.set_option('display.max_columns', 500)\n",
    "\n",
    "#For R (ggplot2)\n",
    "\n",
    "#check the rpy situation by uncommenting the following\n",
    "#%run -m rpy2.situation\n",
    "\n",
    "#use the following on macOS, comment it out on Windows\n",
    "%load_ext rpy2.ipython\n",
    "#on Windows rpy2 doesnt show output of R cells, use the one below if on Windows\n",
    "#use the following on Windows, comment it out on macOS\n",
    "#%load_ext RWinOut "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# DATA\n",
    "\n",
    "The method to generate this map of the research landscape uses publication data from journals and conferences. The clustering is based on the co-citation patterns, i.e. which authors cite each others' papers most often.\n",
    "\n",
    "## Where does the data come from?\n",
    "\n",
    "So as data we need data from publications. The method by Max Noichl used [Web-Of-Science (WOS)](https://www.webofknowledge.com/) as the data source. I cannot publish the data I used to generate the map since I don't have permission to distribute it but I can show you how to get the data on your own.\n",
    "\n",
    "## How to download publication metadata from Web-of-Science (WOS)\n",
    "\n",
    "Open the WOS website and use your institutional sign in to log in. For German universities like the one I am at use the `German Higher Education and Research (DFN-AAI)` option and find your university on the next page.\n",
    "\n",
    "Now let's do a sample search. Enter `International Journal of Architectural Computing` in the search field and change the type to `Publication Name`.\n",
    "\n",
    "![WOS search](images/method/wos_search.png)\n",
    "<img src=\"images/method/wos_search.png\" alt=\"WOS search\" width=\"75%\"/>\n",
    "\n",
    "The search results are 149 as of 5th August 2020.\n",
    "\n",
    "<img src=\"images/method/wos_results.png\" alt=\"WOS results\" width=\"75%\"/>\n",
    "\n",
    "Click on the `Export...` button and choose `Other File Formats`.\n",
    "\n",
    "<img src=\"images/method/wos_export_formats.png\" alt=\"WOS Export formats\" width=\"232\"/>\n",
    "\n",
    "In the next window choose Records from 1 to 149. As content choose `Author, Title, Source, Abstract`. And as File Format choose Plain Text.\n",
    "\n",
    "<img src=\"images/method/wos_export_options.png\" alt=\"WOS Export options\" width=\"290\"/>\n",
    "\n",
    "Click Export to download the file with the data.\n",
    "\n",
    "\n",
    "## How I selected the data about publications on digital technology and architectural design\n",
    "\n",
    "The data used in the next steps has been collected in January 2019. As more publications are being added to WOS over the years, the map will for sure change a bit.\n",
    "\n",
    "### Step 1 — A list of established conferences\n",
    "I made a list of the conferences and journals that where most often discussed at my department ([DDU at TU Darmstadt](https://www.dg.architektur.tu-darmstadt.de/fachgebiet_ddu/index.de.jsp)) and where my colleagues and I have had submitted papers to:\n",
    "\n",
    "Conferences:\n",
    "- AAG\n",
    "- Acadia \n",
    "- CAAD futures\n",
    "- Caadria\n",
    "- Design Modelling Symposium\n",
    "- eCAADe\n",
    "- Fabricate\n",
    "- RobArch\n",
    "- Sigradi\n",
    "- SimAUD\n",
    "\n",
    "Peer-reviewed journals:\n",
    "- IJAC\n",
    "\n",
    "I checked how many papers from those are available on WOS. Not many :(\n",
    "\n",
    "As you see below, I had only 1,600 something publications, yet I needed much more to create the map.\n",
    "Anyway, I downloaded the metadata from WOS."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>SOURCE TITLE</th>\n",
       "      <th>ARTICLES ON WOS</th>\n",
       "      <th>INCLUDE IN MAP</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>ACADIA PROCEEDINGS</td>\n",
       "      <td>209</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>ADVANCES IN ARCHITECTURAL GEOMETRY</td>\n",
       "      <td>22</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>CAAD FUTURES PROCEEDINGS</td>\n",
       "      <td>33</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>CAADRIA PROCEEDINGS</td>\n",
       "      <td>492</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>DESIGN MODELLING SYMPOSIUM PROCEEDINGS</td>\n",
       "      <td>0</td>\n",
       "      <td>no</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>ECAADE PROCEEDINGS</td>\n",
       "      <td>787</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>FABRICATE PROCEEDINGS</td>\n",
       "      <td>0</td>\n",
       "      <td>no</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>INTERNATIONAL JOURNAL OF ARCHITECTURAL COMPUTING</td>\n",
       "      <td>100</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>ROBARCH PROCEEDINGS</td>\n",
       "      <td>0</td>\n",
       "      <td>no</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>SIGRADI PROCEEDINGS</td>\n",
       "      <td>0</td>\n",
       "      <td>no</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>SIMAUD PROCEEDINGS</td>\n",
       "      <td>30</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>- TOTAL -</td>\n",
       "      <td>1673</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                        SOURCE TITLE  ARTICLES ON WOS  \\\n",
       "0                                 ACADIA PROCEEDINGS              209   \n",
       "1                 ADVANCES IN ARCHITECTURAL GEOMETRY               22   \n",
       "2                           CAAD FUTURES PROCEEDINGS               33   \n",
       "3                                CAADRIA PROCEEDINGS              492   \n",
       "4             DESIGN MODELLING SYMPOSIUM PROCEEDINGS                0   \n",
       "5                                 ECAADE PROCEEDINGS              787   \n",
       "6                              FABRICATE PROCEEDINGS                0   \n",
       "7   INTERNATIONAL JOURNAL OF ARCHITECTURAL COMPUTING              100   \n",
       "8                                ROBARCH PROCEEDINGS                0   \n",
       "9                                SIGRADI PROCEEDINGS                0   \n",
       "10                                SIMAUD PROCEEDINGS               30   \n",
       "11                                         - TOTAL -             1673   \n",
       "\n",
       "   INCLUDE IN MAP  \n",
       "0             yes  \n",
       "1             yes  \n",
       "2             yes  \n",
       "3             yes  \n",
       "4              no  \n",
       "5             yes  \n",
       "6              no  \n",
       "7             yes  \n",
       "8              no  \n",
       "9              no  \n",
       "10            yes  \n",
       "11            NaN  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#list the established conferences\n",
    "df = pd.read_csv('data/01a_established_conf.csv', delimiter=';')\n",
    "display(df)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Step 2 — List of authors established in the field\n",
    "So then I decided to start with a list of people and then derive the publication sources from it.\n",
    "I made a list of the authors that are established in the field of digital design in architecture. Then I checked how many articles are listed in WOS for each author and downloaded their metadata."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>LAST NAME</th>\n",
       "      <th>FIRST NAME</th>\n",
       "      <th>ARTICLES ON WOS</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Ahlquist</td>\n",
       "      <td>Sean</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Aish</td>\n",
       "      <td>Robert</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Alexander</td>\n",
       "      <td>Christopher</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Bechthold</td>\n",
       "      <td>Martin</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Brell-Cokcan</td>\n",
       "      <td>Sigrid</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Burry</td>\n",
       "      <td>Mark</td>\n",
       "      <td>34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Burry</td>\n",
       "      <td>Jane</td>\n",
       "      <td>32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Carpo</td>\n",
       "      <td>Mario</td>\n",
       "      <td>34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Cross</td>\n",
       "      <td>Nigel</td>\n",
       "      <td>27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Eisenman</td>\n",
       "      <td>Peter</td>\n",
       "      <td>74</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>Frazer</td>\n",
       "      <td>John</td>\n",
       "      <td>28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>Gehry</td>\n",
       "      <td>Frank</td>\n",
       "      <td>56</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>Gramazio</td>\n",
       "      <td>Fabio</td>\n",
       "      <td>27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>Hensel</td>\n",
       "      <td>Michael</td>\n",
       "      <td>47</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>Hovestadt</td>\n",
       "      <td>Ludger</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>Knight</td>\n",
       "      <td>Terry</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>Kohler</td>\n",
       "      <td>Matthias</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>Kolarevic</td>\n",
       "      <td>Branko</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>Kwinter</td>\n",
       "      <td>Sanford</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>Lynn</td>\n",
       "      <td>Greg</td>\n",
       "      <td>21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>Menges</td>\n",
       "      <td>Achim</td>\n",
       "      <td>80</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>Mitchell</td>\n",
       "      <td>William J.</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>Oxman</td>\n",
       "      <td>Rivka</td>\n",
       "      <td>30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>Picon</td>\n",
       "      <td>Antoine</td>\n",
       "      <td>66</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>Sabin</td>\n",
       "      <td>Jenny</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>Scheurer</td>\n",
       "      <td>Fabian</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>Schumacher</td>\n",
       "      <td>Patrik</td>\n",
       "      <td>23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>Spyropoulos</td>\n",
       "      <td>Theodore</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>Stiny</td>\n",
       "      <td>George</td>\n",
       "      <td>40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>Tamke</td>\n",
       "      <td>Martin</td>\n",
       "      <td>22</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>Tessmann</td>\n",
       "      <td>Oliver</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>Thomsen</td>\n",
       "      <td>Mette Ramsgard</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>Vidler</td>\n",
       "      <td>Anthony</td>\n",
       "      <td>98</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>Weinstock</td>\n",
       "      <td>Michael</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>Woodbury</td>\n",
       "      <td>Robert</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       LAST NAME      FIRST NAME  ARTICLES ON WOS\n",
       "0       Ahlquist            Sean               13\n",
       "1           Aish          Robert               14\n",
       "2      Alexander     Christopher                7\n",
       "3      Bechthold          Martin                9\n",
       "4   Brell-Cokcan          Sigrid               13\n",
       "5          Burry            Mark               34\n",
       "6          Burry            Jane               32\n",
       "7          Carpo           Mario               34\n",
       "8          Cross           Nigel               27\n",
       "9       Eisenman           Peter               74\n",
       "10        Frazer            John               28\n",
       "11         Gehry           Frank               56\n",
       "12      Gramazio           Fabio               27\n",
       "13        Hensel         Michael               47\n",
       "14     Hovestadt          Ludger                4\n",
       "15        Knight           Terry                3\n",
       "16        Kohler        Matthias               19\n",
       "17     Kolarevic          Branko               13\n",
       "18       Kwinter         Sanford               14\n",
       "19          Lynn            Greg               21\n",
       "20        Menges           Achim               80\n",
       "21      Mitchell      William J.                9\n",
       "22         Oxman           Rivka               30\n",
       "23         Picon         Antoine               66\n",
       "24         Sabin           Jenny                7\n",
       "25      Scheurer          Fabian                3\n",
       "26    Schumacher          Patrik               23\n",
       "27   Spyropoulos        Theodore                4\n",
       "28         Stiny          George               40\n",
       "29         Tamke          Martin               22\n",
       "30      Tessmann          Oliver                8\n",
       "31       Thomsen  Mette Ramsgard                8\n",
       "32        Vidler         Anthony               98\n",
       "33     Weinstock         Michael               14\n",
       "34      Woodbury          Robert               14"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#list the selected authors\n",
    "df = pd.read_csv('data/02a_established_authors.csv', delimiter=';')\n",
    "display(df)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If you miss your hero in this list, email me or open an issue in this GitHub repo. However, the purpose of this list is to use it to get to a list of journals, so adding or removing one name might not change the results from the steps that follow.\n",
    "\n",
    "Then I downloaded the publication data for those authors that was available on WOS.\n",
    "\n",
    "\n",
    "Next we are interested in the journals where the authors from the list have published."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RecordCollection glimpse made at: 2020-08-07 19:50:45\n",
      "864 Records from files-from-/Users/antonsavov/_\n",
      "\n",
      "Top Authors\n",
      "1 Menges, Achim\n",
      "2 EISENMAN, P\n",
      "3 Vidler, Anthony\n",
      "4 Carpo, Mario\n",
      "5 Burry, Jane\n",
      "6 PICON, A\n",
      "6 VIDLER, A\n",
      "7 STINY, G\n",
      "7 Picon, Antoine\n",
      "\n",
      "Top Journals\n",
      "1 ARCHITECTURAL DESIGN\n",
      "2 A + U-ARCHITECTURE AND URBANISM\n",
      "3 ARCHITECTURE D AUJOURD HUI\n",
      "4 ENVIRONMENT AND PLANNING B-PLANNING & DESIGN\n",
      "5 DESIGN STUDIES\n",
      "6 ARCHITECTURAL REVIEW\n",
      "7 DIGITAL TURN IN ARCHITECTURE 1992-2012\n",
      "8 LOTUS INTERNATIONAL\n",
      "9 ARCHITECTURAL RECORD\n",
      "\n",
      "Top Cited\n",
      "1 STINY G, 1980, ENVIRON PLANN B, V7, P343, DOI 10.1068/b070343\n",
      "2 Menges A, 2012, ARCHIT DESIGN, V82, P52, DOI 10.1002/ad.1379\n",
      "3 STINY G, 1978, ENVIRON PLANN B, V5, P5, DOI 10.1068/b050005\n",
      "4 Kolarevic Branko., 2003, ARCHITECTURE DIGITAL\n",
      "4 Frazer J., 1995, EVOLUTIONARY ARCHITE\n",
      "4 Gramazio Fabio, 2008, DIGITAL MAT ARCHITEC\n",
      "5 Stiny G, 1975, PICTORIAL FORMAL ASP\n",
      "6 Kolarevic B, 2005, PERFORMATIVE ARCHITE\n",
      "6 STINY G, 1982, ENVIRON PLANN B, V9, P113\n"
     ]
    }
   ],
   "source": [
    "# take a glimpse at the data we have for top authors in the field\n",
    "RCAuthor = mk.RecordCollection(\"data/02b_established_authors\")\n",
    "print(RCAuthor.glimpse())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Step 3 — Get a list of journals where established authors have published\n",
    "\n",
    "I could use the publication data for the top authors to derive a list of journals. I got a list of 210 journals, conference proceedings and books. There were some interesting new sources in the list but also several problems with it. Some sources were not peer-reviewed, others didn't contain publications with citations, yet citing patterns are important for this map, some sources didn't have their focus on architecture and digital technology and so on. So the list needed cleaning up."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Journal</th>\n",
       "      <th>Articles by top authors</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>ARCHITECTURAL DESIGN</td>\n",
       "      <td>191</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>A + U-ARCHITECTURE AND URBANISM</td>\n",
       "      <td>52</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>ARCHITECTURE D AUJOURD HUI</td>\n",
       "      <td>36</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>ENVIRONMENT AND PLANNING B-PLANNING &amp; DESIGN</td>\n",
       "      <td>31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>DESIGN STUDIES</td>\n",
       "      <td>25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>ARCHITECTURAL REVIEW</td>\n",
       "      <td>23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>DIGITAL TURN IN ARCHITECTURE 1992-2012</td>\n",
       "      <td>22</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>LOTUS INTERNATIONAL</td>\n",
       "      <td>21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>ARCHITECTURAL RECORD</td>\n",
       "      <td>18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>NOAH'S ARK: ESSAYS ON ARCHITECTURE</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>AUTOMATION IN CONSTRUCTION</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>COMPUTER-AIDED DESIGN</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>ECAADE 2012, VOL 2: PHYSICAL DIGITALITY</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>ACADIA 2013: ADAPTIVE ARCHITECTURE</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>ARTFORUM INTERNATIONAL</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>INTERNATIONAL JOURNAL OF ARCHITECTURAL COMPUTING</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>COMPUTATIONAL ECOLOGIES: DESIGN IN THE ANTHROP...</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>ACADIA 2014: DESIGN AGENCY</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>ART IN AMERICA</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>PROGRESSIVE ARCHITECTURE</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>SPACE</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>ARTFORUM</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>OPPOSITIONS</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>LEONARDO</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>ORNAMENT: THE POLITICS OF ARCHITECTURE AND SUB...</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>SMART CITIES: A SPATIALISED INTELLIGENCE</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>PROCEEDINGS OF THE 21ST INTERNATIONAL CONFEREN...</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>PROCEEDINGS OF THE 18TH INTERNATIONAL CONFEREN...</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>JAPAN ARCHITECT</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>JOURNAL OF ARCHITECTURAL EDUCATION</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>180</th>\n",
       "      <td>OSIRIS</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>181</th>\n",
       "      <td>THROMBOSIS AND HAEMOSTASIS</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>182</th>\n",
       "      <td>BAUTECHNIK</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>183</th>\n",
       "      <td>OUT OF WATER: DESIGN SOLUTIONS FOR ARID REGIONS</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>184</th>\n",
       "      <td>TENSINET - COST TU1303 INTERNATIONAL SYMPOSIUM...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>185</th>\n",
       "      <td>ARCHIVING 2014, FINAL PROGRAM AND PROCEEDINGS</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>186</th>\n",
       "      <td>COMPUTING SYSTEMS IN ENGINEERING</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>187</th>\n",
       "      <td>POST-PARAMETRIC AUTOMATION IN DESIGN AND CONST...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>188</th>\n",
       "      <td>SIBERIAN HISTORICAL RESEARCH-SIBIRSKIE ISTORIC...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>189</th>\n",
       "      <td>COMPUTER-AIDED ARCHITECTURAL DESIGN: THE NEXT ...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>190</th>\n",
       "      <td>MOUNT SINAI JOURNAL OF MEDICINE</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>191</th>\n",
       "      <td>ARCHITECTURE AND CULTURE</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>192</th>\n",
       "      <td>AMERICAN JOURNAL OF MEDICINE</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>193</th>\n",
       "      <td>JOURNAL OF BUILDING ENGINEERING</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>194</th>\n",
       "      <td>ARCHITECTURE-THE AIA JOURNAL</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>195</th>\n",
       "      <td>JSME INTERNATIONAL JOURNAL SERIES C-MECHANICAL...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>196</th>\n",
       "      <td>ENGINEERING STUDIES</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>197</th>\n",
       "      <td>PLACES-A QUARTERLY JOURNAL OF ENVIRONMENTAL DE...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>198</th>\n",
       "      <td>FUTURE CITY ARCHITECTURE FOR OPTIMAL LIVING</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>199</th>\n",
       "      <td>IEEE CONTROL SYSTEMS MAGAZINE</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>200</th>\n",
       "      <td>APPLIED ACOUSTICS</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>201</th>\n",
       "      <td>COMPUTERS IN INDUSTRY</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>202</th>\n",
       "      <td>NATURE REVIEWS MATERIALS</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>203</th>\n",
       "      <td>SOCIETES</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>204</th>\n",
       "      <td>AMERICAN BEHAVIORAL SCIENTIST</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>205</th>\n",
       "      <td>NEW ADVANCES IN MECHANISMS, MECHANICAL TRANSMI...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>206</th>\n",
       "      <td>JOURNAL OF THE CHEMICAL SOCIETY A -INORGANIC P...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>207</th>\n",
       "      <td>ENVIRONMENT AND PLANNING B-URBAN ANALYTICS AND...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>208</th>\n",
       "      <td>TECHNOLOGY AND CULTURE</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>209</th>\n",
       "      <td>CAADENCE IN ARCHITECTURE: BACK TO COMMAND</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>210 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                               Journal  \\\n",
       "0                                 ARCHITECTURAL DESIGN   \n",
       "1                      A + U-ARCHITECTURE AND URBANISM   \n",
       "2                           ARCHITECTURE D AUJOURD HUI   \n",
       "3         ENVIRONMENT AND PLANNING B-PLANNING & DESIGN   \n",
       "4                                       DESIGN STUDIES   \n",
       "5                                 ARCHITECTURAL REVIEW   \n",
       "6               DIGITAL TURN IN ARCHITECTURE 1992-2012   \n",
       "7                                  LOTUS INTERNATIONAL   \n",
       "8                                 ARCHITECTURAL RECORD   \n",
       "9                   NOAH'S ARK: ESSAYS ON ARCHITECTURE   \n",
       "10                          AUTOMATION IN CONSTRUCTION   \n",
       "11                               COMPUTER-AIDED DESIGN   \n",
       "12             ECAADE 2012, VOL 2: PHYSICAL DIGITALITY   \n",
       "13                  ACADIA 2013: ADAPTIVE ARCHITECTURE   \n",
       "14                              ARTFORUM INTERNATIONAL   \n",
       "15    INTERNATIONAL JOURNAL OF ARCHITECTURAL COMPUTING   \n",
       "16   COMPUTATIONAL ECOLOGIES: DESIGN IN THE ANTHROP...   \n",
       "17                          ACADIA 2014: DESIGN AGENCY   \n",
       "18                                      ART IN AMERICA   \n",
       "19                            PROGRESSIVE ARCHITECTURE   \n",
       "20                                               SPACE   \n",
       "21                                            ARTFORUM   \n",
       "22                                         OPPOSITIONS   \n",
       "23                                            LEONARDO   \n",
       "24   ORNAMENT: THE POLITICS OF ARCHITECTURE AND SUB...   \n",
       "25            SMART CITIES: A SPATIALISED INTELLIGENCE   \n",
       "26   PROCEEDINGS OF THE 21ST INTERNATIONAL CONFEREN...   \n",
       "27   PROCEEDINGS OF THE 18TH INTERNATIONAL CONFEREN...   \n",
       "28                                     JAPAN ARCHITECT   \n",
       "29                  JOURNAL OF ARCHITECTURAL EDUCATION   \n",
       "..                                                 ...   \n",
       "180                                             OSIRIS   \n",
       "181                         THROMBOSIS AND HAEMOSTASIS   \n",
       "182                                         BAUTECHNIK   \n",
       "183    OUT OF WATER: DESIGN SOLUTIONS FOR ARID REGIONS   \n",
       "184  TENSINET - COST TU1303 INTERNATIONAL SYMPOSIUM...   \n",
       "185      ARCHIVING 2014, FINAL PROGRAM AND PROCEEDINGS   \n",
       "186                   COMPUTING SYSTEMS IN ENGINEERING   \n",
       "187  POST-PARAMETRIC AUTOMATION IN DESIGN AND CONST...   \n",
       "188  SIBERIAN HISTORICAL RESEARCH-SIBIRSKIE ISTORIC...   \n",
       "189  COMPUTER-AIDED ARCHITECTURAL DESIGN: THE NEXT ...   \n",
       "190                    MOUNT SINAI JOURNAL OF MEDICINE   \n",
       "191                           ARCHITECTURE AND CULTURE   \n",
       "192                       AMERICAN JOURNAL OF MEDICINE   \n",
       "193                    JOURNAL OF BUILDING ENGINEERING   \n",
       "194                       ARCHITECTURE-THE AIA JOURNAL   \n",
       "195  JSME INTERNATIONAL JOURNAL SERIES C-MECHANICAL...   \n",
       "196                                ENGINEERING STUDIES   \n",
       "197  PLACES-A QUARTERLY JOURNAL OF ENVIRONMENTAL DE...   \n",
       "198        FUTURE CITY ARCHITECTURE FOR OPTIMAL LIVING   \n",
       "199                      IEEE CONTROL SYSTEMS MAGAZINE   \n",
       "200                                  APPLIED ACOUSTICS   \n",
       "201                              COMPUTERS IN INDUSTRY   \n",
       "202                           NATURE REVIEWS MATERIALS   \n",
       "203                                           SOCIETES   \n",
       "204                      AMERICAN BEHAVIORAL SCIENTIST   \n",
       "205  NEW ADVANCES IN MECHANISMS, MECHANICAL TRANSMI...   \n",
       "206  JOURNAL OF THE CHEMICAL SOCIETY A -INORGANIC P...   \n",
       "207  ENVIRONMENT AND PLANNING B-URBAN ANALYTICS AND...   \n",
       "208                             TECHNOLOGY AND CULTURE   \n",
       "209          CAADENCE IN ARCHITECTURE: BACK TO COMMAND   \n",
       "\n",
       "     Articles by top authors  \n",
       "0                        191  \n",
       "1                         52  \n",
       "2                         36  \n",
       "3                         31  \n",
       "4                         25  \n",
       "5                         23  \n",
       "6                         22  \n",
       "7                         21  \n",
       "8                         18  \n",
       "9                         16  \n",
       "10                        15  \n",
       "11                        12  \n",
       "12                         9  \n",
       "13                         9  \n",
       "14                         9  \n",
       "15                         8  \n",
       "16                         8  \n",
       "17                         8  \n",
       "18                         8  \n",
       "19                         8  \n",
       "20                         7  \n",
       "21                         7  \n",
       "22                         7  \n",
       "23                         6  \n",
       "24                         6  \n",
       "25                         6  \n",
       "26                         6  \n",
       "27                         6  \n",
       "28                         5  \n",
       "29                         5  \n",
       "..                       ...  \n",
       "180                        1  \n",
       "181                        1  \n",
       "182                        1  \n",
       "183                        1  \n",
       "184                        1  \n",
       "185                        1  \n",
       "186                        1  \n",
       "187                        1  \n",
       "188                        1  \n",
       "189                        1  \n",
       "190                        1  \n",
       "191                        1  \n",
       "192                        1  \n",
       "193                        1  \n",
       "194                        1  \n",
       "195                        1  \n",
       "196                        1  \n",
       "197                        1  \n",
       "198                        1  \n",
       "199                        1  \n",
       "200                        1  \n",
       "201                        1  \n",
       "202                        1  \n",
       "203                        1  \n",
       "204                        1  \n",
       "205                        1  \n",
       "206                        1  \n",
       "207                        1  \n",
       "208                        1  \n",
       "209                        1  \n",
       "\n",
       "[210 rows x 2 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# list all journal ranked by occurence\n",
    "topJU = RCAuthor.rankedSeries(tag='journal', pandasMode=False)\n",
    "display(pd.DataFrame(topJU,columns=['Journal', 'Articles by top authors']))\n",
    "\n",
    "# And export the list to a CSV file for manual review\n",
    "with open(\"data/03_journals_from_established_authors.csv\", \"w\") as output:\n",
    "    writer = csv.writer(output, lineterminator='\\n')\n",
    "    writer.writerows(topJU)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Step 4 — Clean up the journal list\n",
    "I exported the list of journals where the top authors had published in. Then I checked each journal one by one and excluded the really broad ones such as Architectural Digest, as well those focusing solely on engineering, history, art or aesthethical/political critique. I also removed any books from the list, such as DIGITAL TURN IN ARCHITECTURE by Mario Carpo.\n",
    "\n",
    "Then I merged the list with the established conferences from step 1\n",
    "I was left with 17 Journals.\n",
    "Let's see how many articles this gives us."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>JOURNAL TITLE</th>\n",
       "      <th>ARTICLES BY TOP AUTHORS</th>\n",
       "      <th>ARTICLES ON WOS</th>\n",
       "      <th>INCLUDE</th>\n",
       "      <th>REASON</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3D-ARCH 2015 - 3D VIRTUAL RECONSTRUCTION AND V...</td>\n",
       "      <td>1</td>\n",
       "      <td>63</td>\n",
       "      <td>yes</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>ACADIA PROCEEDINGS</td>\n",
       "      <td>25</td>\n",
       "      <td>209</td>\n",
       "      <td>yes</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>ADVANCES IN ARCHITECTURAL GEOMETRY 2016</td>\n",
       "      <td>5</td>\n",
       "      <td>22</td>\n",
       "      <td>yes</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>AI EDAM-ARTIFICIAL INTELLIGENCE FOR ENGINEERIN...</td>\n",
       "      <td>3</td>\n",
       "      <td>811</td>\n",
       "      <td>yes</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>ARCHITECTURAL DESIGN</td>\n",
       "      <td>191</td>\n",
       "      <td>6585</td>\n",
       "      <td>yes</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>AUTOMATION IN CONSTRUCTION</td>\n",
       "      <td>15</td>\n",
       "      <td>2325</td>\n",
       "      <td>yes</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>CAAD FUTURES PROCEEDINGS</td>\n",
       "      <td>1</td>\n",
       "      <td>33</td>\n",
       "      <td>yes</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>CAADRIA PROCEEDINGS</td>\n",
       "      <td>19</td>\n",
       "      <td>492</td>\n",
       "      <td>yes</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>COMPUTER-AIDED DESIGN</td>\n",
       "      <td>12</td>\n",
       "      <td>3811</td>\n",
       "      <td>yes</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>DESIGN STUDIES</td>\n",
       "      <td>25</td>\n",
       "      <td>506</td>\n",
       "      <td>yes</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>ECAADE PROCEEDINGS</td>\n",
       "      <td>35</td>\n",
       "      <td>787</td>\n",
       "      <td>yes</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>ENVIRONMENT AND PLANNING B-PLANNING &amp; DESIGN</td>\n",
       "      <td>31</td>\n",
       "      <td>2836</td>\n",
       "      <td>yes</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>INTERNATIONAL JOURNAL OF ARCHITECTURAL COMPUTING</td>\n",
       "      <td>8</td>\n",
       "      <td>100</td>\n",
       "      <td>yes</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>MASS CUSTOMISATION AND PERSONALISATION IN ARCH...</td>\n",
       "      <td>1</td>\n",
       "      <td>22</td>\n",
       "      <td>yes</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>NEXUS NETWORK JOURNAL</td>\n",
       "      <td>4</td>\n",
       "      <td>423</td>\n",
       "      <td>yes</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>PERFORMANCE-ORIENTED ARCHITECTURE: RETHINKING ...</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "      <td>yes</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>SIMAUD PROCEEDINGS</td>\n",
       "      <td>1</td>\n",
       "      <td>30</td>\n",
       "      <td>yes</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>- - -</td>\n",
       "      <td>- - -</td>\n",
       "      <td>- - -</td>\n",
       "      <td>- - -</td>\n",
       "      <td>- - -</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>3D PRINTING AND ADDITIVE MANUFACTURING</td>\n",
       "      <td>4</td>\n",
       "      <td>167</td>\n",
       "      <td>no</td>\n",
       "      <td>too broad</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>A + U-ARCHITECTURE AND URBANISM</td>\n",
       "      <td>52</td>\n",
       "      <td>8906</td>\n",
       "      <td>no</td>\n",
       "      <td>too broad</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>ADVANCES IN ENGINEERING SOFTWARE</td>\n",
       "      <td>5</td>\n",
       "      <td>2657</td>\n",
       "      <td>no</td>\n",
       "      <td>focus on engineering</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>ARCHITECTURAL RECORD</td>\n",
       "      <td>18</td>\n",
       "      <td>14553</td>\n",
       "      <td>no</td>\n",
       "      <td>too broad</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>ARCHITECTURAL REVIEW</td>\n",
       "      <td>23</td>\n",
       "      <td>11429</td>\n",
       "      <td>no</td>\n",
       "      <td>too broad</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>ARCHITECTURE</td>\n",
       "      <td>5</td>\n",
       "      <td>4012</td>\n",
       "      <td>no</td>\n",
       "      <td>too broad</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>ARCHITECTURE D AUJOURD HUI</td>\n",
       "      <td>36</td>\n",
       "      <td>6721</td>\n",
       "      <td>no</td>\n",
       "      <td>too broad</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>DESIGN ISSUES</td>\n",
       "      <td>3</td>\n",
       "      <td>1076</td>\n",
       "      <td>no</td>\n",
       "      <td>too broad</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>DESIGN QUARTERLY</td>\n",
       "      <td>3</td>\n",
       "      <td>281</td>\n",
       "      <td>no</td>\n",
       "      <td>not technologically focused</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>DIGITAL TURN IN ARCHITECTURE 1992-2012</td>\n",
       "      <td>22</td>\n",
       "      <td>23</td>\n",
       "      <td>no</td>\n",
       "      <td>book, single author</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>DOMUS</td>\n",
       "      <td>3</td>\n",
       "      <td>5445</td>\n",
       "      <td>no</td>\n",
       "      <td>too broad</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>ENVIRONMENT AND PLANNING B-URBAN ANALYTICS AND...</td>\n",
       "      <td>1</td>\n",
       "      <td>145</td>\n",
       "      <td>no</td>\n",
       "      <td>not architecturally focused</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>HARVARD ARCHITECTURE REVIEW</td>\n",
       "      <td>4</td>\n",
       "      <td>78</td>\n",
       "      <td>no</td>\n",
       "      <td>too broad</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>JAPAN ARCHITECT</td>\n",
       "      <td>5</td>\n",
       "      <td>2487</td>\n",
       "      <td>no</td>\n",
       "      <td>too broad</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>JOURNAL OF ARCHITECTURAL EDUCATION</td>\n",
       "      <td>5</td>\n",
       "      <td>1466</td>\n",
       "      <td>no</td>\n",
       "      <td>too broad</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>LOTUS INTERNATIONAL</td>\n",
       "      <td>21</td>\n",
       "      <td>1999</td>\n",
       "      <td>no</td>\n",
       "      <td>too broad</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>NOAH'S ARK: ESSAYS ON ARCHITECTURE</td>\n",
       "      <td>16</td>\n",
       "      <td>17</td>\n",
       "      <td>no</td>\n",
       "      <td>too broad</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>OPPOSITIONS</td>\n",
       "      <td>7</td>\n",
       "      <td>94</td>\n",
       "      <td>no</td>\n",
       "      <td>too broad</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>PROGRESSIVE ARCHITECTURE</td>\n",
       "      <td>8</td>\n",
       "      <td>4592</td>\n",
       "      <td>no</td>\n",
       "      <td>too broad</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>ROUTLEDGE COMPANION FOR ARCHITECTURE DESIGN AN...</td>\n",
       "      <td>3</td>\n",
       "      <td>46</td>\n",
       "      <td>no</td>\n",
       "      <td>too broad</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                        JOURNAL TITLE ARTICLES BY TOP AUTHORS  \\\n",
       "0   3D-ARCH 2015 - 3D VIRTUAL RECONSTRUCTION AND V...                       1   \n",
       "1                                  ACADIA PROCEEDINGS                      25   \n",
       "2             ADVANCES IN ARCHITECTURAL GEOMETRY 2016                       5   \n",
       "3   AI EDAM-ARTIFICIAL INTELLIGENCE FOR ENGINEERIN...                       3   \n",
       "4                                ARCHITECTURAL DESIGN                     191   \n",
       "5                          AUTOMATION IN CONSTRUCTION                      15   \n",
       "6                            CAAD FUTURES PROCEEDINGS                       1   \n",
       "7                                 CAADRIA PROCEEDINGS                      19   \n",
       "8                               COMPUTER-AIDED DESIGN                      12   \n",
       "9                                      DESIGN STUDIES                      25   \n",
       "10                                 ECAADE PROCEEDINGS                      35   \n",
       "11       ENVIRONMENT AND PLANNING B-PLANNING & DESIGN                      31   \n",
       "12   INTERNATIONAL JOURNAL OF ARCHITECTURAL COMPUTING                       8   \n",
       "13  MASS CUSTOMISATION AND PERSONALISATION IN ARCH...                       1   \n",
       "14                              NEXUS NETWORK JOURNAL                       4   \n",
       "15  PERFORMANCE-ORIENTED ARCHITECTURE: RETHINKING ...                       1   \n",
       "16                                 SIMAUD PROCEEDINGS                       1   \n",
       "17                                              - - -                   - - -   \n",
       "18             3D PRINTING AND ADDITIVE MANUFACTURING                       4   \n",
       "19                    A + U-ARCHITECTURE AND URBANISM                      52   \n",
       "20                   ADVANCES IN ENGINEERING SOFTWARE                       5   \n",
       "21                               ARCHITECTURAL RECORD                      18   \n",
       "22                               ARCHITECTURAL REVIEW                      23   \n",
       "23                                       ARCHITECTURE                       5   \n",
       "24                         ARCHITECTURE D AUJOURD HUI                      36   \n",
       "25                                      DESIGN ISSUES                       3   \n",
       "26                                   DESIGN QUARTERLY                       3   \n",
       "27             DIGITAL TURN IN ARCHITECTURE 1992-2012                      22   \n",
       "28                                              DOMUS                       3   \n",
       "29  ENVIRONMENT AND PLANNING B-URBAN ANALYTICS AND...                       1   \n",
       "30                        HARVARD ARCHITECTURE REVIEW                       4   \n",
       "31                                    JAPAN ARCHITECT                       5   \n",
       "32                 JOURNAL OF ARCHITECTURAL EDUCATION                       5   \n",
       "33                                LOTUS INTERNATIONAL                      21   \n",
       "34                 NOAH'S ARK: ESSAYS ON ARCHITECTURE                      16   \n",
       "35                                        OPPOSITIONS                       7   \n",
       "36                           PROGRESSIVE ARCHITECTURE                       8   \n",
       "37  ROUTLEDGE COMPANION FOR ARCHITECTURE DESIGN AN...                       3   \n",
       "\n",
       "   ARTICLES ON WOS INCLUDE                       REASON  \n",
       "0               63     yes                          NaN  \n",
       "1              209     yes                          NaN  \n",
       "2               22     yes                          NaN  \n",
       "3              811     yes                          NaN  \n",
       "4             6585     yes                          NaN  \n",
       "5             2325     yes                          NaN  \n",
       "6               33     yes                          NaN  \n",
       "7              492     yes                          NaN  \n",
       "8             3811     yes                          NaN  \n",
       "9              506     yes                          NaN  \n",
       "10             787     yes                          NaN  \n",
       "11            2836     yes                          NaN  \n",
       "12             100     yes                          NaN  \n",
       "13              22     yes                          NaN  \n",
       "14             423     yes                          NaN  \n",
       "15               9     yes                          NaN  \n",
       "16              30     yes                          NaN  \n",
       "17           - - -   - - -                        - - -  \n",
       "18             167      no                    too broad  \n",
       "19            8906      no                    too broad  \n",
       "20            2657      no         focus on engineering  \n",
       "21           14553      no                    too broad  \n",
       "22           11429      no                    too broad  \n",
       "23            4012      no                    too broad  \n",
       "24            6721      no                    too broad  \n",
       "25            1076      no                    too broad  \n",
       "26             281      no  not technologically focused  \n",
       "27              23      no          book, single author  \n",
       "28            5445      no                    too broad  \n",
       "29             145      no  not architecturally focused  \n",
       "30              78      no                    too broad  \n",
       "31            2487      no                    too broad  \n",
       "32            1466      no                    too broad  \n",
       "33            1999      no                    too broad  \n",
       "34              17      no                    too broad  \n",
       "35              94      no                    too broad  \n",
       "36            4592      no                    too broad  \n",
       "37              46      no                    too broad  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#the updated list of journals\n",
    "df = pd.read_csv('data/04a_cleaned_up_journals.csv', delimiter=';')\n",
    "display(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RecordCollection glimpse made at: 2020-08-07 19:50:51\n",
      "19031 Records from files-from-/Users/antonsavov/_\n",
      "\n",
      "Top Authors\n",
      "1 [Anonymous]\n",
      "2 TOY, M\n",
      "3 JENCKS, C\n",
      "4 Toy, M\n",
      "5 Castle, H\n",
      "6 Castle, Helen\n",
      "7 COOKE, C\n",
      "8 Menges, Achim\n",
      "9 Batty, Michael\n",
      "\n",
      "Top Journals\n",
      "1 ARCHITECTURAL DESIGN\n",
      "2 COMPUTER-AIDED DESIGN\n",
      "3 ENVIRONMENT AND PLANNING B-PLANNING & DESIGN\n",
      "4 AUTOMATION IN CONSTRUCTION\n",
      "5 AI EDAM-ARTIFICIAL INTELLIGENCE FOR ENGINEERING DESIGN ANALYSIS AND MANUFACTUR\n",
      "6 DESIGN STUDIES\n",
      "7 NEXUS NETWORK JOURNAL\n",
      "8 PROCEEDINGS OF THE 19TH INTERNATIONAL CONFERENCE ON COMPUTER-AIDED ARCHITECTUR\n",
      "9 PROCEEDINGS OF THE 20TH INTERNATIONAL CONFERENCE ON COMPUTER-AIDED ARCHITECTUR\n",
      "\n",
      "Top Cited\n",
      "1 STINY G, 1980, ENVIRON PLANN B, V7, P409, DOI 10.1068/b070409\n",
      "2 Schon D, 1983, REFLECTIVE PRACTITIO\n",
      "3 STINY G, 1978, ENVIRON PLANN B, V5, P5, DOI 10.1068/b050005\n",
      "4 Hillier B, 1984, SOCIAL LOGIC SPACE\n",
      "5 Piegl L., 1997, NURBS BOOK\n",
      "6 GERO JS, 1990, AI MAG, V11, P26\n",
      "7 Goldberg D. E, 1989, GENETIC ALGORITHMS S\n",
      "8 Eastman C, 2011, BIM HDB GUIDE BUILDI\n",
      "9 RITTEL HWJ, 1973, POLICY SCI, V4, P155, DOI 10.1007/BF01405730\n"
     ]
    }
   ],
   "source": [
    "# take a glimpse at the data we have for now\n",
    "RCtmp = mk.RecordCollection(\"data/04b_cleaned_up_journals\")\n",
    "print(RCtmp.glimpse())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A little over 19000 articles. Not bad. I wanted to check for more.\n",
    "\n",
    "### Step 5 — Fish for more data using Scimago\n",
    "\n",
    "So using the website [Scimago](https://www.scimagojr.com/journalrank.php?category=2216) and filtering to the field of architecture, I looked for sources that:\n",
    "1. Had a high ranking\n",
    "2. Focused on digital technology and architecture\n",
    "3. Had data available on WOS\n",
    "\n",
    "This resulted only in one new journal being added to the list:\n",
    "\n",
    "TECHNE-JOURNAL OF TECHNOLOGY FOR ARCHITECTURE AND ENVIRONMENT\n",
    "\n",
    "However the whole exersize was quite interesting to get an overview of the publication landscape in architecture. If one day I decide to make a map of the whole architecture research, not just the one relating to digital design, the insights would be useful.\n",
    "\n",
    "So let's load the data for the map."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "### Step 6 — load and clean up data\n",
    "\n",
    "First let's load the data and take a glimpse of what we have so far."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RecordCollection glimpse made at: 2020-08-07 19:50:57\n",
      "19317 Records from files-from-/Users/antonsavov/_\n",
      "\n",
      "Top Authors\n",
      "1 [Anonymous]\n",
      "2 TOY, M\n",
      "3 JENCKS, C\n",
      "4 Toy, M\n",
      "5 Castle, H\n",
      "6 Castle, Helen\n",
      "7 COOKE, C\n",
      "8 Menges, Achim\n",
      "9 Batty, Michael\n",
      "\n",
      "Top Journals\n",
      "1 ARCHITECTURAL DESIGN\n",
      "2 COMPUTER-AIDED DESIGN\n",
      "3 ENVIRONMENT AND PLANNING B-PLANNING & DESIGN\n",
      "4 AUTOMATION IN CONSTRUCTION\n",
      "5 AI EDAM-ARTIFICIAL INTELLIGENCE FOR ENGINEERING DESIGN ANALYSIS AND MANUFACTUR\n",
      "6 DESIGN STUDIES\n",
      "7 NEXUS NETWORK JOURNAL\n",
      "8 TECHNE-JOURNAL OF TECHNOLOGY FOR ARCHITECTURE AND ENVIRONMENT\n",
      "9 PROCEEDINGS OF THE 19TH INTERNATIONAL CONFERENCE ON COMPUTER-AIDED ARCHITECTUR\n",
      "\n",
      "Top Cited\n",
      "1 STINY G, 1980, ENVIRON PLANN B, V7, P409, DOI 10.1068/b070409\n",
      "2 Schon D, 1983, REFLECTIVE PRACTITIO\n",
      "3 STINY G, 1978, ENVIRON PLANN B, V5, P5, DOI 10.1068/b050005\n",
      "4 Hillier B, 1984, SOCIAL LOGIC SPACE\n",
      "5 Piegl L., 1997, NURBS BOOK\n",
      "6 GERO JS, 1990, AI MAG, V11, P26\n",
      "7 Goldberg D. E, 1989, GENETIC ALGORITHMS S\n",
      "8 Eastman C, 2011, BIM HDB GUIDE BUILDI\n",
      "9 RITTEL HWJ, 1973, POLICY SCI, V4, P155, DOI 10.1007/BF01405730\n"
     ]
    }
   ],
   "source": [
    "date_string = datetime.datetime.now().strftime(\"%Y-%m-%d-%H:%M\")\n",
    "\n",
    "#read the data for all journals\n",
    "RC = mk.RecordCollection(\"data/06_all_journals\")\n",
    "print(RC.glimpse())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "An intersting thing we can do with the data at this point is to see which are the TOP30 journals where the articles cited in the articles in the collection are published. This can help find more journals to include in the dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Journal</th>\n",
       "      <th>Times cited</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>COMPUT AIDED DESIGN</td>\n",
       "      <td>2681</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>AUTOMAT CONSTR</td>\n",
       "      <td>1922</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>ENVIRON PLANN B</td>\n",
       "      <td>1604</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>ACM T GRAPHIC</td>\n",
       "      <td>981</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>J CONSTR ENG M ASCE</td>\n",
       "      <td>848</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>J COMPUT CIVIL ENG</td>\n",
       "      <td>829</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>IEEE COMPUT GRAPH</td>\n",
       "      <td>779</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>DESIGN STUDIES</td>\n",
       "      <td>778</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>COMPUT AIDED GEOM D</td>\n",
       "      <td>774</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>LECT NOTES COMPUT SC</td>\n",
       "      <td>713</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>THESIS</td>\n",
       "      <td>655</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>IEEE T PATTERN ANAL</td>\n",
       "      <td>610</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>COMMUN ACM</td>\n",
       "      <td>605</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>AI EDAM</td>\n",
       "      <td>601</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>DESIGN STUD</td>\n",
       "      <td>591</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>ADV ENG INFORM</td>\n",
       "      <td>589</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>ARTIF INTELL</td>\n",
       "      <td>587</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>J MECH DESIGN</td>\n",
       "      <td>567</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>ENVIRON PLANN A</td>\n",
       "      <td>525</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>COMPUT GRAPH FORUM</td>\n",
       "      <td>519</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>J CONSTR ENG M</td>\n",
       "      <td>517</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>RES ENG DES</td>\n",
       "      <td>488</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>COMPUTER-AIDED GEOMETRIC DESIGN</td>\n",
       "      <td>429</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>INT J PROD RES</td>\n",
       "      <td>427</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>IEEE T VIS COMPUT GR</td>\n",
       "      <td>426</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>INT J ADV MANUF TECH</td>\n",
       "      <td>422</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>BUILD ENVIRON</td>\n",
       "      <td>422</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>COMPUT IND</td>\n",
       "      <td>415</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>COMPUT GRAPH</td>\n",
       "      <td>408</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>COMPUTATIONAL GEOMET</td>\n",
       "      <td>383</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                            Journal  Times cited\n",
       "0               COMPUT AIDED DESIGN         2681\n",
       "1                    AUTOMAT CONSTR         1922\n",
       "2                   ENVIRON PLANN B         1604\n",
       "3                     ACM T GRAPHIC          981\n",
       "4               J CONSTR ENG M ASCE          848\n",
       "5                J COMPUT CIVIL ENG          829\n",
       "6                 IEEE COMPUT GRAPH          779\n",
       "7                    DESIGN STUDIES          778\n",
       "8               COMPUT AIDED GEOM D          774\n",
       "9              LECT NOTES COMPUT SC          713\n",
       "10                           THESIS          655\n",
       "11              IEEE T PATTERN ANAL          610\n",
       "12                       COMMUN ACM          605\n",
       "13                          AI EDAM          601\n",
       "14                      DESIGN STUD          591\n",
       "15                   ADV ENG INFORM          589\n",
       "16                     ARTIF INTELL          587\n",
       "17                    J MECH DESIGN          567\n",
       "18                  ENVIRON PLANN A          525\n",
       "19               COMPUT GRAPH FORUM          519\n",
       "20                   J CONSTR ENG M          517\n",
       "21                      RES ENG DES          488\n",
       "22  COMPUTER-AIDED GEOMETRIC DESIGN          429\n",
       "23                   INT J PROD RES          427\n",
       "24             IEEE T VIS COMPUT GR          426\n",
       "25             INT J ADV MANUF TECH          422\n",
       "26                    BUILD ENVIRON          422\n",
       "27                       COMPUT IND          415\n",
       "28                     COMPUT GRAPH          408\n",
       "29             COMPUTATIONAL GEOMET          383"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# list TOP30 journals from the cited works in these articles\n",
    "\n",
    "topJUcited = []\n",
    "for R in RC:\n",
    "    #topJUcited.append(list(set(R.getCitations().get('citeString'))))\n",
    "    topJUcited.append(list(set(R.getCitations().get(\"journal\"))))\n",
    "#print(topJUcited[:10])\n",
    "rankedJU = {}\n",
    "for jlist in topJUcited:\n",
    "    for ju in jlist:\n",
    "        rankedJU[ju] = rankedJU.get(ju, 0) + 1\n",
    "        \n",
    "sorted_by_value = sorted(rankedJU.items(), key=lambda kv: kv[1], reverse=True)            \n",
    "#RC.rankedSeries(tag='journal', pandasMode=False)\n",
    "display(pd.DataFrame(sorted_by_value[:30], columns=['Journal','Times cited']))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Next we will remove the articles which have less then a certain number of citations to reduce the noise in the map. We can also remove all articles that don't list the year of their publication, or are without abstracts. This helps later when we want to understand what each cluster is researching about. Let's print a glimpse of the data so far."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RecordCollection glimpse made at: 2020-08-07 19:50:59\n",
      "12572 Records from Empty\n",
      "\n",
      "Top Authors\n",
      "1 [Anonymous]\n",
      "2 Menges, Achim\n",
      "3 Li, Heng\n",
      "4 Batty, Michael\n",
      "5 BATTY, M\n",
      "6 Wang, Xiangyu\n",
      "7 JENCKS, C\n",
      "8 Batty, M\n",
      "8 Gero, John S.\n",
      "\n",
      "Top Journals\n",
      "1 COMPUTER-AIDED DESIGN\n",
      "2 AUTOMATION IN CONSTRUCTION\n",
      "3 ENVIRONMENT AND PLANNING B-PLANNING & DESIGN\n",
      "4 ARCHITECTURAL DESIGN\n",
      "5 AI EDAM-ARTIFICIAL INTELLIGENCE FOR ENGINEERING DESIGN ANALYSIS AND MANUFACTUR\n",
      "6 DESIGN STUDIES\n",
      "7 NEXUS NETWORK JOURNAL\n",
      "8 TECHNE-JOURNAL OF TECHNOLOGY FOR ARCHITECTURE AND ENVIRONMENT\n",
      "9 PROCEEDINGS OF THE 18TH INTERNATIONAL CONFERENCE ON COMPUTER-AIDED ARCHITECTUR\n",
      "\n",
      "Top Cited\n",
      "1 STINY G, 1980, ENVIRON PLANN B, V7, P409, DOI 10.1068/b070409\n",
      "2 Schon D, 1983, REFLECTIVE PRACTITIO\n",
      "3 STINY G, 1978, ENVIRON PLANN B, V5, P5, DOI 10.1068/b050005\n",
      "4 Hillier B, 1984, SOCIAL LOGIC SPACE\n",
      "5 Piegl L., 1997, NURBS BOOK\n",
      "6 GERO JS, 1990, AI MAG, V11, P26\n",
      "7 Goldberg D. E, 1989, GENETIC ALGORITHMS S\n",
      "8 Eastman C, 2011, BIM HDB GUIDE BUILDI\n",
      "9 RITTEL HWJ, 1973, POLICY SCI, V4, P155, DOI 10.1007/BF01405730\n"
     ]
    }
   ],
   "source": [
    "#prune the collection from articles that miss key data\n",
    "RC2 = mk.RecordCollection()\n",
    "\n",
    "for R in RC:\n",
    "    randnr = randint(0, 4)\n",
    "    \n",
    "    if len(R.getCitations().get(\"author\"))>=3: # and randnr==0 apply condition in order to downsample records\n",
    "        #Here we kick out every paper that cites less then 3 authors. Why? because they\n",
    "        #are so dissimilar from the others, that they only produce noise.  \n",
    "   \n",
    "        try:\n",
    "            R['year']\n",
    "            #if R['year']>1961:\n",
    "            #R['abstract']  #Add this when working with abstracts. It removes every paper that has none. \n",
    "            #This can sometimes remove whole journals, that are archived without abstracts, so handle with care.\n",
    "            RC2.add(R)\n",
    "        except KeyError:\n",
    "            pass\n",
    "    else:\n",
    "        pass\n",
    "    \n",
    "\n",
    "print(RC2.glimpse())\n",
    "\n",
    "\n",
    "RC = RC2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Ok, we lost 7,000 articles. The clean up reduced the amount of datapoints to about 12,000 which is still ok. \n",
    "\n",
    "The glimpse shows us the most prolific researchers where we have the computational design superstar Achim Menges. We see the journals with the most publications in the dataset. And we see the top cited articles where the winner is [George Stiny's \"Kindergarten Grammars: Designing with Froebel's Building Gifts\"](https://journals.sagepub.com/doi/abs/10.1068/b070409).The [The Palladian Grammar](https://journals.sagepub.com/doi/abs/10.1068/b050005) again by Stiny is 3rd. [Bill Hillier's Space Syntax Book](https://www.cambridge.org/core/books/social-logic-of-space/6B0A078C79A74F0CC615ACD8B250A985) and [Horst Rittel's Dillemas in a General Theory of Planning](https://link.springer.com/article/10.1007/BF01405730) also make it in the TOP 10. So far so good.\n",
    "\n",
    "Next let's assemble a clean list of all included journals and conferences so we can print it next to the map."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 228,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "THE MAP IS GENERATED FROM DATA FROM THE FOLLOWING PUBLICATIONS:\n",
      "3D-ARCH 2015 - 3D VIRTUAL RECONSTRUCTION AND VISUALIZATION\n",
      "    OF COMPLEX ARCHITECTURES (63)\n",
      "AUTOMATION IN CONSTRUCTION (2273)\n",
      "ARCHITECTURAL DESIGN (1683)\n",
      "AI EDAM-ARTIFICIAL INTELLIGENCE FOR ENGINEERING DESIGN\n",
      "    ANALYSIS AND MANUFACTURING (734)\n",
      "ADVANCES IN ARCHITECTURAL GEOMETRY 2016 (22)\n",
      "ACADIA CONFERENCE PROCEEDINGS (118)\n",
      "COMPUTER-AIDED DESIGN (3168)\n",
      "COMPUTATIONAL ECOLOGIES: DESIGN IN THE ANTHROPOCENE (44)\n",
      "CAADRIA CONFERENCE PROCEEDINGS (447)\n",
      "CAAD FUTURES CONFERENCE PROCEEDINGS (33)\n",
      "DESIGN STUDIES (486)\n",
      "ENVIRONMENT AND PLANNING B-PLANNING & DESIGN (2030)\n",
      "ECAADE CONFERENCE PROCEEDINGS (731)\n",
      "INTERNATIONAL JOURNAL OF ARCHITECTURAL COMPUTING (85)\n",
      "MASS CUSTOMISATION AND PERSONALISATION IN ARCHITECTURE AND\n",
      "    CONSTRUCTION (17)\n",
      "NEXUS NETWORK JOURNAL (369)\n",
      "PERFORMANCE-ORIENTED ARCHITECTURE: RETHINKING ARCHITECTURAL\n",
      "    DESIGN AND THE BUILT ENVIRONMENT (7)\n",
      "SIMAUD CONFERENCE PROCEEDINGS (29)\n",
      "TECHNE-JOURNAL OF TECHNOLOGY FOR ARCHITECTURE AND\n",
      "    ENVIRONMENT (233)\n",
      "DATA SNAPSHOT 23 JAN 2019\n"
     ]
    }
   ],
   "source": [
    "#list all journals where the papers in the data were published\n",
    "# list all journal ranked by occurence\n",
    "topJU = RC.rankedSeries(tag='journal', pandasMode=False)\n",
    "#conference proceedings or journals sometimes appear as separate journals for each year or issue. lets combine them\n",
    "combineConfProc = [\n",
    "    # list of what to search for and what to change the name to and tally it up as\n",
    "    [['acadia'], 'ACADIA CONFERENCE PROCEEDINGS'],\n",
    "    [['ecaade','fusion:'], 'ECAADE CONFERENCE PROCEEDINGS'],\n",
    "    [['caadria'], 'CAADRIA CONFERENCE PROCEEDINGS'],\n",
    "    [['simaud'], 'SIMAUD CONFERENCE PROCEEDINGS'],\n",
    "    [['caad futures'], 'CAAD FUTURES CONFERENCE PROCEEDINGS'],\n",
    "]\n",
    "journalList = []\n",
    "confprocdict = {}\n",
    "for j in topJU:\n",
    "    #if conf name found, combine in conferences\n",
    "    isConf = False\n",
    "    jouName = j[0]\n",
    "    for conf in combineConfProc:\n",
    "        for searchphrase in conf[0]:\n",
    "            if searchphrase.lower() in jouName.lower():\n",
    "                isConf = True\n",
    "                if not conf[1] in confprocdict:\n",
    "                    confprocdict[conf[1]] = [conf[1],j[1]]\n",
    "                else:\n",
    "                    confprocdict[conf[1]][1] = confprocdict[conf[1]][1] + j[1]\n",
    "    #wrapepdJuName = textwrap.fill(j[0],70)\n",
    "    if not isConf:\n",
    "        aju = [ jouName, j[1]]\n",
    "        journalList.append(aju)\n",
    "    #print(wrapepdJuName + \" (\" + str(j[1])+ \")\")\n",
    "    #print()\n",
    "for conf in confprocdict:\n",
    "    journalList.append(confprocdict[conf])\n",
    "\n",
    "wrapper=textwrap.TextWrapper()\n",
    "wrapper.width=60\n",
    "wrapper.initial_indent=\"\"\n",
    "wrapper.subsequent_indent=\"    \"\n",
    "    \n",
    "#sortedJournalList\n",
    "journalList = sorted(journalList, key=lambda x: x[0][0])\n",
    "journalListLabels = []\n",
    "for journal in journalList:\n",
    "    label = wrapper.fill(journal[0] + \" (\" + str(journal[1]) + \")\")#+\"\\n\"\n",
    "    journalListLabels.append(label)\n",
    "    #print(label )\n",
    "journalListLabel = \"THE MAP IS GENERATED FROM DATA FROM THE FOLLOWING PUBLICATIONS:\" + \"\\n\"+\"\\n\".join([w for w in journalListLabels]) + \"\\nDATA SNAPSHOT 23 JAN 2019\"\n",
    "print(journalListLabel)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# MAP\n",
    "\n",
    "Now let’s do the map. The method is explained below with the code. The code for the first 4 steps (extracting the fetures, dimensionality reduction, 2D embedding with umap and clustering) is largely based on the [code shared by Max Noichl here](https://homepage.univie.ac.at/maximilian.noichl/posts/structure-of-recent-philosophy-ii/), so refer to his Jupyter Notebook for details.\n",
    "\n",
    "## Build a multi-dimensional space of relationships\n",
    "\n",
    "In order to arrange the publications in a 2D map (which is done with UMAP), and subsequently to determine the clusters, the publications datapoints need to be assigned `features`. The datapoints with more similar features will be placed closer to each other. Max Noichl carried out some experiments with features such as author, words in abstracts, title and so on and he recommends using the following two features:\n",
    "- the cited works\n",
    "- the cited authors\n",
    "\n",
    "To extract the features we use `scikit-learn` and clean up the weakly linked points.\n",
    "\n",
    "First let's do the cited works, then the cited authors and then fitler out weak links."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "########### Cited Works - Features ############\n",
    "\n",
    "from sklearn.feature_extraction.text import TfidfVectorizer, CountVectorizer\n",
    "import re\n",
    "drc = pd.DataFrame.from_dict(RC.forNLP(extraColumns=['journal','AU','FU','PD']))\n",
    "\n",
    "d = []\n",
    "citedAU = []\n",
    "citestring =[]\n",
    "for R in RC:\n",
    "\n",
    "    d.append(list(set(R.getCitations().get(\"citeString\")))) #To cluster by cited author\n",
    "    citedAU.append(list(set(R.getCitations().get(\"author\"))))\n",
    "    citestring.append(list(set(R.getCitations().get(\"citeString\"))))\n",
    "\n",
    "drc[\"citedAU\"] = citedAU\n",
    "drc[\"citestring\"] = citestring\n",
    "#print(d[0])\n",
    "authorslist = ['§'.join(filter(None,x)) for x in list(d)] \n",
    "#print(authorslist[0])\n",
    "\n",
    "# vec = TfidfVectorizer(token_pattern=r'(?<=[^|§])[\\s\\w,\\.:;]+(?=[$|§])')\n",
    "vec = CountVectorizer(token_pattern=r'(?<=[§])[\\s\\w,\\.:;\\/\\[\\]-]+(?=[§])',binary=True, min_df = 3)#, min_df = 1)\n",
    "\n",
    "\n",
    "Xrc = vec.fit_transform(authorslist)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "########### Authors - Features ############\n",
    "\n",
    "d = []\n",
    "for R in RC:\n",
    "    authors = list(set(R.getCitations().get(\"author\")))\n",
    "#    print(authors)\n",
    "    authors = filter(None, authors)\n",
    "    f = []\n",
    "    for a in authors:\n",
    "        f.append(' '.join([w for w in a.split(' ')if len(w)>2]))\n",
    "        \n",
    "    authors = f#' '.join(f)\n",
    "    d.append(authors)\n",
    "authorslist = [';'.join(filter(None,x)) for x in list(d)] \n",
    "vec = CountVectorizer(token_pattern=r'(?<=[;])[\\s\\w]+(?=[;])',binary=True, min_df = 10)\n",
    "\n",
    "XrcAu = vec.fit_transform(authorslist)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "k = [XrcAu,Xrc]\n",
    "XrcFull = scipy.sparse.hstack(k).tocsr()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "###### Filtering #######\n",
    "from scipy.sparse import coo_matrix, vstack\n",
    "from scipy.sparse import csr_matrix\n",
    "import scipy as scipy\n",
    "row_names = np.array(drc[\"id\"])\n",
    "\n",
    "newdf=[]\n",
    "a = 0\n",
    "# index by name:\n",
    "for x in range(0,XrcFull.shape[0]): #Xrc.shape[0]):\n",
    "    row_idx, = np.where(row_names == drc[\"id\"][x])\n",
    "    if np.diff(XrcFull[row_idx].tocsr().indptr) >= 4:\n",
    "        if a == 0:\n",
    "            k = [XrcFull[row_idx]]\n",
    "        if a != 0:\n",
    "            k.append(XrcFull[row_idx])\n",
    "        a = a+1\n",
    "        newdf.append(drc.loc[x])\n",
    "        \n",
    "drc = pd.DataFrame(newdf).reset_index()\n",
    "M = scipy.sparse.vstack((k))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Preliminary dimensionality reduction with SVD\n",
    "\n",
    "Here is what Max Noichl says about this next step:\n",
    "> \"This is strictly speaking not necessary: We could pass our vectors directly to the umap-algorithm. But when we use a lot of data, we can get a sharper image when we clear out some noise with SVD beforehand. If we were interested more in classification and less in visualization, I would suggest to skip this step, reduce with umap to ~thirty dimensions and cluster on that.\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.37508746334120097\n"
     ]
    }
   ],
   "source": [
    "from sklearn.decomposition import TruncatedSVD\n",
    "SVD = TruncatedSVD(n_components=350, n_iter=7, random_state=42)\n",
    "\n",
    "XSVD = SVD.fit_transform(M)\n",
    "print(SVD.explained_variance_ratio_.sum())\n",
    "dSVD = pd.DataFrame(XSVD)\n",
    "\n",
    "sSVD = dSVD[[0,1]]\n",
    "sSVD.columns = ['x','y']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Project to 2D space\n",
    "\n",
    "To generate a 2D projection we creating a 2D-embedding with [UMAP](https://github.com/lmcinnes/umap).\n",
    "\n",
    "As far as I understand the algorithm here, each feature established above is a dimension along which distance between articles is measured. So we have a space of several hundred dimensions. How to plot that, i.e. project it to a 2D space so we can vizualize it as a map?\n",
    "\n",
    "`UMAP` let's us reduce the dimensions to two and is good for two reasons: speed and preserving relationships.\n",
    "\n",
    "So in the next step we use `UMAP` to create a 2D-embedding of the data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "#if this cells issues a numba warning, ignore it.\n",
    "#the umap routine should have completed anyway.\n",
    "#more about the issue: https://github.com/lmcinnes/umap/issues/252\n",
    "\n",
    "import umap\n",
    "\n",
    "try:\n",
    "    drc = drc.drop('x',axis=1)\n",
    "    drc = drc.drop('y',axis=1)\n",
    "\n",
    "except KeyError:\n",
    "    pass\n",
    "\n",
    "n_neighbors = 7\n",
    "\n",
    "embedding = umap.UMAP(n_neighbors = n_neighbors,#small => local, large => global: 5-50\n",
    "                      min_dist = 0.0005, #small => local, large => global: 0.001-0.5\n",
    "                      spread = 1.5,\n",
    "                      metric='cosine').fit_transform(XSVD)\n",
    "embedding = pd.DataFrame(embedding)\n",
    "embedding.columns = ['x','y']\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "And next let's plot it to see how it looks like."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 279,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABLAAAAMgCAYAAAAz4JsCAAAEGWlDQ1BrQ0dDb2xvclNwYWNlR2VuZXJpY1JHQgAAOI2NVV1oHFUUPrtzZyMkzlNsNIV0qD8NJQ2TVjShtLp/3d02bpZJNtoi6GT27s6Yyc44M7v9oU9FUHwx6psUxL+3gCAo9Q/bPrQvlQol2tQgKD60+INQ6Ium65k7M5lpurHeZe58853vnnvuuWfvBei5qliWkRQBFpquLRcy4nOHj4g9K5CEh6AXBqFXUR0rXalMAjZPC3e1W99Dwntf2dXd/p+tt0YdFSBxH2Kz5qgLiI8B8KdVy3YBevqRHz/qWh72Yui3MUDEL3q44WPXw3M+fo1pZuQs4tOIBVVTaoiXEI/MxfhGDPsxsNZfoE1q66ro5aJim3XdoLFw72H+n23BaIXzbcOnz5mfPoTvYVz7KzUl5+FRxEuqkp9G/Ajia219thzg25abkRE/BpDc3pqvphHvRFys2weqvp+krbWKIX7nhDbzLOItiM8358pTwdirqpPFnMF2xLc1WvLyOwTAibpbmvHHcvttU57y5+XqNZrLe3lE/Pq8eUj2fXKfOe3pfOjzhJYtB/yll5SDFcSDiH+hRkH25+L+sdxKEAMZahrlSX8ukqMOWy/jXW2m6M9LDBc31B9LFuv6gVKg/0Szi3KAr1kGq1GMjU/aLbnq6/lRxc4XfJ98hTargX++DbMJBSiYMIe9Ck1YAxFkKEAG3xbYaKmDDgYyFK0UGYpfoWYXG+fAPPI6tJnNwb7ClP7IyF+D+bjOtCpkhz6CFrIa/I6sFtNl8auFXGMTP34sNwI/JhkgEtmDz14ySfaRcTIBInmKPE32kxyyE2Tv+thKbEVePDfW/byMM1Kmm0XdObS7oGD/MypMXFPXrCwOtoYjyyn7BV29/MZfsVzpLDdRtuIZnbpXzvlf+ev8MvYr/Gqk4H/kV/G3csdazLuyTMPsbFhzd1UabQbjFvDRmcWJxR3zcfHkVw9GfpbJmeev9F08WW8uDkaslwX6avlWGU6NRKz0g/SHtCy9J30o/ca9zX3Kfc19zn3BXQKRO8ud477hLnAfc1/G9mrzGlrfexZ5GLdn6ZZrrEohI2wVHhZywjbhUWEy8icMCGNCUdiBlq3r+xafL549HQ5jH+an+1y+LlYBifuxAvRN/lVVVOlwlCkdVm9NOL5BE4wkQ2SMlDZU97hX86EilU/lUmkQUztTE6mx1EEPh7OmdqBtAvv8HdWpbrJS6tJj3n0CWdM6busNzRV3S9KTYhqvNiqWmuroiKgYhshMjmhTh9ptWhsF7970j/SbMrsPE1suR5z7DMC+P/Hs+y7ijrQAlhyAgccjbhjPygfeBTjzhNqy28EdkUh8C+DU9+z2v/oyeH791OncxHOs5y2AtTc7nb/f73TWPkD/qwBnjX8BoJ98VQNcC+8AAAA4ZVhJZk1NACoAAAAIAAGHaQAEAAAAAQAAABoAAAAAAAKgAgAEAAAAAQAABLCgAwAEAAAAAQAAAyAAAAAAr6+39AAAQABJREFUeAHsnQeAJFXVts/kuDmTc46KiIqIEoyggphQEVRAwmciGH5QEVFUBEFUQBFBEDOIiiKCIkgSUJaw5Lhs3p3d2cnpP2+tNVbXVM90np7p58Bsd1e4de9zb1dXvXXOuVXr1q0bMgwCEIAABCAAAQhAAAIQgAAEIAABCEAAAmVKoLpM60W1IAABCEAAAhCAAAQgAAEIQAACEIAABCAQEEDAYiBAAAIQgAAEIAABCEAAAhCAAAQgAAEIlDUBBKyy7h4qBwEIQAACEIAABCAAAQhAAAIQgAAEIICAxRiAAAQgAAEIQAACEIAABCAAAQhAAAIQKGsCCFhl3T1UDgIQgAAEIAABCEAAAhCAAAQgAAEIQAABizEAAQhAAAIQgAAEIAABCEAAAhCAAAQgUNYEELDKunuoHAQgAAEIQAACEIAABCAAAQhAAAIQgAACFmMAAhCAAAQgAAEIQAACEIAABCAAAQhAoKwJIGCVdfdQOQhAAAIQgAAEIAABCEAAAhCAAAQgAAEELMYABCAAAQhAAAIQgAAEIAABCEAAAhCAQFkTQMAq6+6hchCAAAQgAAEIQAACEIAABCAAAQhAAAIIWIwBCEAAAhCAAAQgAAEIQAACEIAABCAAgbImgIBV1t1D5SAAAQhAAAIQgAAEIAABCEAAAhCAAAQQsBgDEIAABCAAAQhAAAIQgAAEIAABCEAAAmVNAAGrrLuHykEAAhCAAAQgAAEIQAACEIAABCAAAQggYDEGIAABCEAAAhCAAAQgAAEIQAACEIAABMqaAAJWWXcPlYMABCAAAQhAAAIQgAAEIAABCEAAAhBAwGIMQAACEIAABCAAAQhAAAIQgAAEIAABCJQ1AQSssu4eKgcBCEAAAhCAAAQgAAEIQAACEIAABCCAgMUYgAAEIAABCEAAAhCAAAQgAAEIQAACEChrAghYZd09VA4CEIAABCAAAQhAAAIQgAAEIAABCEAAAYsxAAEIQAACEIAABCAAAQhAAAIQgAAEIFDWBBCwyrp7qBwEIAABCEAAAhCAAAQgAAEIQAACEIAAAhZjAAIQgAAEIAABCEAAAhCAAAQgAAEIQKCsCSBglXX3UDkIQAACEIAABCAAAQhAAAIQgAAEIAABBCzGAAQgAAEIQAACEIAABCAAAQhAAAIQgEBZE0DAKuvuoXIQgAAEIAABCEAAAhCAAAQgAAEIQAACCFiMAQhAAAIQgAAEIAABCEAAAhCAAAQgAIGyJoCAVdbdQ+UgAAEIQAACEIAABCAAAQhAAAIQgAAEELAYAxCAAAQgAAEIQAACEIAABCAAAQhAAAJlTQABq6y7h8pBAAIQgAAEIAABCEAAAhCAAAQgAAEIIGAxBiAAAQhAAAIQgAAEIAABCEAAAhCAAATKmgACVll3D5WDAAQgAAEIQAACEIAABCAAAQhAAAIQQMBiDEAAAhCAAAQgAAEIQAACEIAABCAAAQiUNQEErLLuHioHAQhAAAIQgAAEIAABCEAAAhCAAAQggIDFGIAABCAAAQhAAAIQgAAEIAABCEAAAhAoawIIWGXdPVQOAhCAAAQgAAEIQAACEIAABCAAAQhAAAGLMQABCEAAAhCAAAQgAAEIQAACEIAABCBQ1gQQsMq6e6gcBCAAAQhAAAIQgAAEIAABCEAAAhCAAAIWYwACEIAABCAAAQhAAAIQgAAEIAABCECgrAkgYJV191A5CEAAAhCAAAQgAAEIQAACEIAABCAAAQQsxgAEIAABCEAAAhCAAAQgAAEIQAACEIBAWRNAwCrr7qFyEIAABCAAAQhAAAIQgAAEIAABCEAAAghYjAEIQAACEIAABCAAAQhAAAIQgAAEIACBsiaAgFXW3UPlIAABCEAAAhCAAAQgAAEIQAACEIAABBCwGAMQgAAEIAABCEAAAhCAAAQgAAEIQAACZU0AAausu4fKQQACEIAABCAAAQhAAAIQgAAEIAABCCBgMQYgAAEIQAACEIAABCAAAQhAAAIQgAAEypoAAlZZdw+VgwAEIAABCEAAAhCAAAQgAAEIQAACEEDAYgxAAAIQgAAEIAABCEAAAhCAAAQgAAEIlDUBBKyy7h4qBwEIQAACEIAABCAAAQhAAAIQgAAEIICAxRiAAAQgAAEIQAACEIAABCAAAQhAAAIQKGsCCFhl3T1UDgIQgAAEIAABCEAAAhCAAAQgAAEIQAABizEAAQhAAAIQgAAEIAABCEAAAhCAAAQgUNYEELDKunuoHAQgAAEIQAACEIAABCAAAQhAAAIQgEAtCCYGgeXLl9tjjz2WUtktttjCNt1005RllfRh1apV9sgjj6Q0eeONN7atttoqZdlE/1Ap7Zyo/bR27Vp78MEHU6o/f/5823bbbVOW8QECEIAABCAAAQhAAAIQgAAEcieAgJU7u5Lueeutt9rHPvaxlGOeeeaZdsopp6QsK9cPa9assRdffNFmz55tCxYsKEg1//nPf9qRRx6ZUtZJJ51k55xzTsqyif5hsrbzhRdesCuvvNLuvPNOO+SQQ+x973ufTZ06dcJ118MPP2xvfetbU+p99NFH23e+852UZXyAAAQgAAEIQAACEIAABCAAgdwJIGDlzm7UPSXWHHTQQaNuk8nKW265pWCCTybHK+Q27e3t9qMf/cguu+wyk1gRWmtrq+288852xhln2H777Rcu5rXCCHzkIx+xu+66K2j1bbfdZs8//7x99atfrTAKpWnuQw89ZEcccUTKweTBeeONN6YsCz9IkHv66afDj8HrddddZ9tvv/3wsh133HH4ffSNznsXXnhhdNGI90888YQdeuihI5ZrwQknnGAnn3xy4rrowt7eXnvVq15lnZ2dw4sljuucmYmlq3+4b0NDQ+BFp+123XVXe8c73mF1dXXhal4hAAEIQAACEIAABCAAgRITQMAqEvCBgQFbvHhx3qWrnIloTz31lL373e823ajGbf369Xb33Xfb2972tsCD6rvf/a7V1NTEN+PzBCTQ399vfX19KTWvr68f0b/PPffcsHgVbvzzn/8cASuEUeBXiT3x81FjY2PaoyxdunTE9vF+jZcXFnbDDTcE3mdVVVXhohGvf/vb30aUH260bt268O2orxI94+cX1WnhwoWB4DTqzr4yXf2j+0nE+/Of/xwsOv/88+0HP/iB7bbbbtFNeA8BCEAAAhCAAAQgAAEIlIgASdxLBLqSDrN69Wp705veNOLmMonB1VdfbZdccknSKpZNQAIXX3yxzZs3L+XvmmuuGdGSjTbayGbOnJmyfJdddkn5zIeJSUA52+TxNZpJwMrXfv/73ycWkW554sZZLFSb9t9/f0saz1kUw6YQgAAEIAABCEAAAhCAQI4EELByBMdu6QmcdtpptmzZshEbbLnlliZvnLidffbZJo8PrHIIKBTr85//vE2fPj1o9GabbWaf+tSnKgfAJG/paALV4OCg3X777XkRGBoasj/84Q+JZcgDrFgmD8NTTz3VlixZUqxDUC4EIAABCEAAAhCAAAQgkIYAIYRpwOS7eO7cuXbttdeOKEZ5oeLJ2HUzf9VVV43YVguU9Hwi2UsvvWS//OUvU6qsxNzXX3+9vfzlLw+EqgMOOCAlJ5ZCCpWkXkm8scohcOyxx9pRRx1lSoK+55572mghZ5VDZXK09O9//3vaPFaasVGTOuRj//rXvxJFcpUpT6lnn33WlOMrW/vpT39qtbUbfhYVdqlQZ4UNRkO5dQ6XiKVtMQhAAAIQgAAEIAABCECgdAQQsIrEuqmpyd7ylreMKF3hNXFT/qekbePbTYTPv/71r03eEVGTUCHxSjZ//nx7+9vfbsp7FbV4LpvoOt5PXgJKlP2yl71s8jawQlummTPlrRSKQVEMErfytbHCBLVeM5Jmawp9jnqJKnH7XnvtZZpVMmq/+93vbMWKFTZnzpzoYt5DAAIQgAAEIAABCEAAAkUkgIBVRLilKFqeAcoj9Y9//MOeeeYZ22mnnex1r3udHX744WMe/vHHHzclQpZHxCOPPBJ4eykP0R577BEIatXV2UeYypsmbm9+85tTFu29994pn/XhySefHLEsXPCf//zHbrrpJpPXRUdHh+2zzz4jbijDbfN9VXLxuOfcO9/5Tps1a5ZdccUVQT023XRT++EPf5h4qHyY5tNOcY/f1GsciFVoixYtCjzhws96feUrXxnk9YkuC99LaLjvvvsCjxaNLYX5aXypP+OJrJXo+t///veIxOwq649//KPJM0/hgscdd1xQvMLIvvnNb4aHCl7F+KMf/WjKsuiHxx57zFQncVJbNOOc6iHvrQMPPDDRgyuJi7y+lKfrN7/5jUloUXmbbLJJwELekUmiS7Qemb6XB49m4bzjjjsCsUP1lBii8RSftECCyKOPPppStGbYi8/SedFFF6XMuqcd3v/+95vGZDmZvCr1fY2Ov7B+5SxghXWMvupceu655wZjLrpcYxABK0qE9xCAAAQgAAEIQAACECguAQSs4vItaumareuII46wm2++efg499xzTyC0/PWvf7ULLrggxZtgeCN/c9lll9lnP/vZETPGSWyQveENbwhuviUqZGPyMIvPbqaE3VGTWBC3qNdDdJ1mpvv4xz8eeHOEyyW6XXnllfahD30oXFSwVwlYX/3qV1PKk1CixM0SO2QSUpIsH6b5tlNhU/F6SySJCggSSOLbfPKTnxwhYEl8UD4q1SlqErN++9vf2te//nX7zGc+Y6effrop/FX2pz/9KRgv0e3D98pVpD/lQIsKWPG6bL/99mkFLHnsnXnmmSnjQOVL+JEdfPDBwWQA8fGaxGX33XcPRF+FtYam740ELdVToWFhbq5wfbavbW1tgaehxKvQ7rzzzuDtr371q+A7Gv2ePP/88yP6RrN0RgWslStX2he+8IWwuOBVYtsJJ5yQsqxcPkioio4/1UszGYbfo1zrKZE47rF52GGHBf0XlnnXXXeZeBUqBFsepBKsoqbPr33ta6OLeA8BCEAAAhCAAAQgAAEIFJFA9i42RawMRWdH4Dvf+U6KeBXdWzfh8v5IshNPPDEQIHQzmc5uueUWe81rXmNJIY/p9tFy5b9avnx5yt/GG2+csouEkLgleZBILJFHjEKR4qak79/4xjfii4vy+Zxzzkm56U7K1ZQP03Jpp+BJFJXnVly8ioKV15/YK1l/KUweU0r4njQOwuPLQ2/fffcd1ZMv3FYiWlS8CpfrVeKovlf5mvo0Kl5Fy5NIrDZFTQJc3OLfk1AAi24nL60pU6ZEF43b+2222Sbl2EmJ3OWV1dnZObxdfJ/hFaO8iSdvl/j9uc99LmUPefiFYnzKihw/xGfMVDFMPJEjTHaDAAQgAAEIQAACEIBAjgQQsHIEVw67xXNNxev0ve99LyX5sNbLMyFdwvj4/gr7+v73vx9fnPfnJA+MJE+G888/P+9jFaIAcYhaPPwrX6bl0k618ZJLLhnh3RJte/T9j3/84yC8MLqs0O8VFikxKBNbvHixfelLXxpz07ESiF9++eUpIsuYBeawwY033mjyJAptu+22C0I0w8961Ux30dnukgSsJOErWkYp3yucs7W1dfiQ9957r3V1dQ1/1pu4qCWRPFuLh8oqJFnee/Gk7fHtsj1OdPv4OUDrtt566+gmvIcABCAAAQhAAAIQgAAEikwAAavIgItdvELAlLPpXe9614i8OgqHU16iqH3ta1+LfgzeKwRJN9SXXnrpiJAbCRryyimUKddV3INCMzbK6ydq8j6J112eT5r9S6FeCn+MC0nR/Yv5Pn7cfJiWUzs1u1o8ub4mIzjvvPNMOa4+8YlPpGCVl8u3vvWtYJk8pNQWjce4ffGLXwzWheF+8fXpPqv8r3zlKyNWv/e977UbbrghyKGl+kVNx3jggQeiixLfy7NP3xvljwrDIMMNJXApL1y+pgTgP/vZz4Lv1bbbbjuiOIXBRu2ggw6Kfgzey2MptCTht5wELHl0vv71rw+ra8oBFhfd4vmv5DWXjcnrScJY1Pbff//gY/wcoplNFQ6br+n8J4/UuO2www7xRXyGAAQgAAEIQAACEIAABIpIgBxYRYRb7KKVG+qss84KDqNZDJVTJz61+4svvjg8A6BytuimLmpvfOMbg5xG4TIlbo8m0l67dm2QuFwCWSFMApRubKOm8Lu4KCSBIm4nn3yynXHGGcFiJe2WeCFxpBSmmRPFe8cdd0y5Kc6XaTm1U+JP3DvplFNOCcI4xVjhahKyormAwqT9Smatv3gOKu0ngTJJwNG60Uy53eKJzSV4SFSVmCnBQuFzxx9/fEoxSnQuL6p0JgFMopws/N7Et1dOqnj+pnTlJS2XR+EVV1xh4UQImmlRCdyjJpZnn3328CIJWPGwXwlYhxxySDB5gZLNR0255DQey8U0wYJyUUXHtDyulE9PptDBqPikSQH0l40lhQWGwpWErJ/85CfDxfX09AQh1hISc7WnnnrKPv3pT4/4XujcI68vDAIQgAAEIAABCEAAAhAoHQEErNKxLviR4jNgKdFwXMBSWFVo0ZClcJlujqMW9aAIl+tmvhB23XXXmZLLR005cCRgxS0pv0z8RjST/DkqOy7KRI915JFH2lvf+tboohHvFRolYSFMND9jxozhbfJlWqh2DlcojzfPPvvsiL0POOCAlGUSYaIC1tNPPx2EqcYFyJSdcvwQPU5YhDymojnIJJhIZIt62iTtF+6v16222ir6MZh1M2WBf3jhhRfii7L6rLEZilfaUSGCSmKv2RxDi/e9hBiNsajAG3pg6VW5x6ImEbecTAJW3Iss6nElb6xo3j1tq32ysXhYoARMnfdkoZAVLU/bx88b0fXx9xob4fhSP8RDIMPtlUy/paUl/MgrBCAAAQhAAAIQgAAEIFACAghYJYBcqkNEhZXwmNEk7AopjJvyxkRzaSlZsW68Fb4VWiEELOWQkSdD3OQJEwpD0XXxm3utiwsP0e3Tvf/LX/4yarJl5c8Zy3SDnFRH7Zcv00K1c6w2ZLI+SbSZN29eyq5KgC5voqjpRj8eyhddn+v7JEFtjz32SClOXocK5QqFHq1M6pOUnWIfpk6dGltiWQsrIwpIWKDJDKIClrwb5ZXU3NwcbC1B5NWvfnVKniiFQ+q7mBQ+KO/JcjK1RTN2SvANQzAVBiwBWecmeWNFTfWPJnSPrkt6r1C+qCCmbZRDSzMxyjTj4C677JKSl01ebhLN4mGiwQ4J/2QSLq3zQZLonlAciyAAAQhAAAIQgAAEIACBAhJAwCogzPEuKvQciNYjKk4lCRRve9vbopsnvk/aL3HDNAslcHzgAx8IprWPbqJZ7JI8vrRNkrAzffr06O5l8T6JTTZMy6mdSUJlnPnuu+9u+iuFJQlYEiniFg9bVC4vCbfx5fH9ws9J35twXSFfk4Qy9X9UmJVXUlTokWfZY489NiKXlASZJI+jQtY327JCbyrl5QoFLJ1/br/99iAMMio+NTQ02H777Zd2NsikY0uMjnpwaZs4A4URPvTQQ8O7SyT8xz/+MRzGOLwixzfyWFXet2J4HOZYJXaDAAQgAAEIQAACEIBAxRCorpiW0lBbtmxZThR0E5iPSaiKesioLOXFUeLvdNbW1paySp420ZCslJXj+CFfpuXUzqRQy/G8UV+xYsWIno3OcheuTArlSto33H68XpO81ELRJ6xTklfV3XffnZI7StsqH1kSi7Cc8XgN2xJvg4Qrja1oDi/lCJPnWbhPJvWNhw9qH50TlCst/NN5Im7RnFzxdZl8lleqPON++ctf2tVXXx14mWWyH9tAAAIQgAAEIAABCEAAAoUlgAdWYXlSWozAVVddNSKh9vz584OcUqMJUvF18fw/scOk/fjtb3971DClUnkTpatgodqZrvxslsfron2jHnzZlFWIbeWlE7f+/v74osQ6ZhoyNqKwEi+I11O5spTYPOoN9+Mf/3iE0BPPNVXiaiceLgwHfMUrXmESfVavXh1sJ48yeUpFx1IocmUqYMmL86abbhpx3NNPP33EsvgCzXqq80AmnnbXX3/9cLihwoa33nrrjD354sflMwQgAAEIQAACEIAABCBQWAIIWIXlWdalKT9N3D772c+OORNYPFl8vIx0n5X/Jinv1Q9+8IMxbwo1c51mUAxNoUPd3d3BTIvhskxeMwnny6ScdNvky7RQ7UxXv2yWK0fTwoULU3bRTG5JnkMpGxXpg4Sce+65J6V0efLEwxqjCdy1sYQKzdBXbhYKPNF6JX23JE5FZyNUHqy4hQJQfHm6z0neWurbdBYP1dN2SYJidP9QjJIQquT/8liSaaKDa6+9NrqpKcxQFu6TsjLhw2233WYKDc3FFKYpD1AJa2OZcmqly3c31r6shwAEIAABCEAAAhCAAASKSwABq7h8y6r0aK6dsGI777yzvf3tbw8/FuxVOYg0w1/8Jvn444/PKB9NkjAk8SJpecEqnUNB+TJNak8u7UwK9UvyVhqtiUroHzeJQ1HBSGFg8XBQzQxYDJFr8803j1fHlixZEszmF10RT9oupmOJLdH9S/U+9EgKj6c+k6dS3OICVny9xDklrs/G1LcSlqKTMyxfvjxtEfHQWCVK1yyKo5nKDhOmS2ALBSztEw3j23bbbYfLkiidiSWFD2ayX7iN9s9EwAq35xUCEIAABCAAAQhAAAIQKD8C5MAqvz4pWo0UDhO3pLCccJuoB1S4LJNX3cgec8wxFk9wvv3229tZZ52VSREmb6C43XXXXfFF4/45X6aFaqfCMuMWJtIOl0fFi3BZ9DVJMFq0aFF0E9OskZ/61KeG/77whS+MKRZl6mWTciD/kCSYxMerxugTTzyRsmtSO1I2GIcPCoGVJ1LU5H2VFNamcLt4aGF0v1zCB+VVFOeisLy4GKnjyHOyq6sresgg0Xw421/KitiHMExQHlhJbdPm2XqPqUyFAcZN/JL+kpL35yuAxY/NZwhAAAIQgAAEIAABCECg9AQQsErPfNyOuOuuu44Ij7nmmmtSvCNUOSVt/+QnP2l77rnniJvuTCovkerWW29N2VQ3v5dcckkw5b08g5L+wptf7bjjjjum7K8PF1100XDIkTy7Lr300hHblHpBvkwL1c4kIezOO++0xYsXB0jk/XPuueeOiuflL3/5iPXf/OY3Lcw/9vTTT1t0JjltrPpHc2fNmDFjRBkKA4z27YgN0iyQkBMXTZQPKhSCJMideeaZI8rOReBJU4WcF//xj3+0qGeYkn/Lsy5q6byolJReoWzpLAy/S7c+3fKksXbqqaem5NvSeNGyuMlTMxuTiLT33nsn7pJt/SWyxT3CJMY99dRTiX8SNOOJ/bUsHDeJlWIhBCAAAQhAAAIQgAAEIFD2BBCwyr6LCldB5Vs6+eSTUwqUOKFQP4kXhx12WCBabbrppkHidYlEJ554YkrYUcrOCR9uv/32IGFyfJUEq/333z8ImVLYVNLfX/7yl+HdjjjiiBE3obqR1Y30m9/8ZlMYUlxMGd65hG/yZVqodkrAmjJlSkrLFcapJPUHHnigKTl43JsqZWP/sM8++5hmh4uaRDB5QkkUkqAZF2FUdtSSQip/9atfBf227777Rjcd871EiqOOOiplOx1f5WgM7LbbbqayoybhRGGq420SXPbaa68g19N+++034nun+ul7l87SiXDyzJKwl4sp311UbFQZ9913n+2yyy62zTbbBN8piVya9TBqCnX8zGc+E12U0fskoUq5uDSjXzaW5D01GgOJnkljLRrGmM3x2RYCEIAABCAAAQhAAAIQKA8CCFjl0Q8lq8Upp5xiSeFm8lDQVPTyaoiabma1PFN79tlnM9101O2mTp1q733ve0dsI0+iO+64w9ra2kasG68F+TAtVDsVInb00UePQKAwMXlA6TUTO+OMM0ZsJtYaB3EvKoVPxoUNCYzNzc0jylCo32OPPTZi+VgLTjvttBGJ+5U3SWMgOlNfWI7qk5SwPFxfylcJwAp7VUhenJ36/dBDD01bnSTxRxtL/Mm1fXvssYcdd9xxicdUPqy4l1O44UknnRSIheHnTF+T2iARO9sk6UkClsoZzZIErqRyRiuDdRCAAAQgAAEIQAACEIBAeRFAwCqv/ih6bRRa8+tf/zrwiMnkYCeccMLwjGGZbF/IbT7/+c8HHkTpyvz4xz+eblVJl+fLtFDtVF/Nnj07bdszyT0kL6zzzz9/zLxWyj2kkNDGxsaU4ymEUIJHksW9f5K2iS9TQnZ5ziSFSEa3Vb6lz33ucyYG5WDvec970uaxkhfVxRdfPGrie+WLkydk3NJ5ZsW3S/dZAuUHP/jBdKtHLNe24pqLyUMuLpZnMgajx1LYXzzHmdYnCVTR/ZIELnmbvfTSS9HNeA8BCEAAAhCAAAQgAAEITCACCFgTqLMKVVXlbdK09PJuSRIGNttsM3vnO99pSpj99a9/vVCHzbociSQ33nhj4KkS9dqQh4/qrpxO8+bNy7rcYuyQD9NCtXOjjTYK+lWha1GTwKZcUZdffnl0cdr3H/nIR4LwTIlZ8dn8JL4o7FFeXelyHJ1++ulB/8RnRszVc+iVr3ylKTT18MMPt3iOLYWLSSi5/vrrA6ElF5EsLYg8Vig88Oc//7kpxDRqCsf8zW9+k9HMn0liVZJXU7T8sd6rDySe6bstrnEBUvurz7VOnpfaNsmjbqzjaL1ExXgbsq1/UvL2nXbaKUjePlod5AmYlMw9qbzRymEdBCAAAQhAAAIQgAAEIFA+BKrWrVs3VD7VoSbjQcDHQJAfqb293RRmlHTjNx71ih5TIXALFy4MQuEk0Iw2S1t0v/F6nyvTQrVz5cqVplkI5ZGlm/m4mJQpF+VIC5NlKyeVco9lyl7jSWGDysUljyLtn25mukzro+0UjvjII48E3j1KhB4VN7Mpp1TbyotIdVY+smy+W1/72tdMf6HJI+vhhx8OPxbkVaGNqpv6WEnxlQtrk002GZErqyAHoxAIQAACEIAABCAAAQhAAAJ5EEDAygMeu0IAAhAoBgEJS/KAe/TRR4eLP+aYY+yCCy4Y/swbCEAAAhCAAAQgAAEIQAAClUSAEMJK6m3aCgEITAgCP/3pT1PEK1U62/C7CdFQKgkBCEAAAhCAAAQgAAEIQCBDAnhgZQiKzSAAAQgUm8Cpp54a5B9btGhRyqGUJ03LMg3fTNmZDxCAAAQgAAEIQAACEIAABCYBgdpJ0AaaAAEIQGBSEFAy+qVLl45oy4c+9CHEqxFUWAABCEAAAhCAAAQgAAEIVBIBQggrqbdpKwQgMOEIKFH9KaecMuHqTYUhAAEIQAACEIAABCAAAQgUkgACViFpUhYEIACBAhJ49atfbb/4xS+spaWlgKVSFAQgAAEIQAACEIAABCAAgYlHgBxYE6/PqDEEIDBJCfznP/+xhx56yPr6+mynnXayvffee5K2lGZBAAIQgAAEIAABCEAAAhDIjgACVna82BoCEIAABCAAAQhAAAIQgAAEIAABCECgxAQIISwxcA4HAQhAAAIQgAAEIAABCEAAAhCAAAQgkB0BBKzseLE1BCAAAQhAAAIQgAAEIAABCEAAAhCAQIkJIGCVGDiHgwAEIAABCEAAAhCAAAQgAAEIQAACEMiOAAJWdrzYGgIQgAAEIAABCEAAAhCAAAQgAAEIQKDEBBCwSgycw0EAAhCAAAQgAAEIQAACEIAABCAAAQhkRwABKztebA0BCEAAAhCAAAQgAAEIQAACEIAABCBQYgIIWCUGzuEgAAEIQAACEIAABCAAAQhAAAIQgAAEsiOAgJUdL7aGAAQgAAEIQAACEIAABCAAAQhAAAIQKDEBBKwSA+dwEIAABCAAAQhAAAIQgAAEIAABCEAAAtkRQMDKjhdbQwACEIAABCAAAQhAAAIQgAAEIAABCJSYAAJWiYFzOAhAAAIQgAAEIAABCEAAAhCAAAQgAIHsCCBgZceLrSEAAQhAAAIQgAAEIAABCEAAAhCAAARKTAABq8TAORwEIAABCEAAAhCAAAQgAAEIQAACEIBAdgQQsLLjxdYQgAAEIACBvAn09/fnXUa5FDAwMFAuVaEe40RgMo3nydSWcRoOHBYCEIAABCBQNAIIWEVDS8EQgAAEIACBkQR6enrs/e9/vz399NMjV07AJX/729/stNNOm4A1p8qFIHDLLbfYpz/96UIUVRZlfOhDH7JFixaVRV2oBAQgAAEIQAACqQQQsFJ5lO2nj3zkI3b00UeXbf1yrdgVV1xhW265Za67j/t+usidN2+ePfjgg+Nel1JVoFh9FrJ8+OGHS9UUjgOBEQTOOecc22effUYsL+SCz33uczY0NGSbb775iGK7u7vt0UcftfXr149Yl27BmjVr7Lnnnku3Ou3ywcHBtOuiK3p7e4Mb+ra2tuji4fevec1r7F//+pddfPHFw8smwpudd97Zvve975VlVZcuXRqMkVwrF55PM/1tUh/rt+zXv/51Vod89tlng2uTd7/73Yn7ZTueNSYff/xx6+rqSixvtIWZjmexfeKJJyzd9u95z3vssMMOs/b29tEOxzoIQAACEIAABMaBAALWOEDP5ZB6Yq8LwclmfX19OV2olgsHXQDrQjvdhXC51LOQ9ShWn40Hy5/97Gc2e/ZsO/bYYzNG9Ja3vCXY5x//+EfG+yRt+H//93920EEHJa2a9MtC7jqvjWZf+cpXbLfddhttk4Kv0418Mc+1d9xxh/3mN7+xSy65xGpqaobrv2LFCnvXu95lCxYssFe+8pW20UYb2Zve9CZ78sknh7eJv1m4cKG9+tWvDoSwXXfd1bbffnu76qqr4puN+Kyx+973vte+9rWvjVgXXbB27Vr72Mc+Fggbe++9t2222Wa23377WVxkbmxsDNrz9a9/3Z5//vloEWX9vrOz03Q+y8Re9rKX2Rve8IYxRaVNNtnELrvsskyKTLvNiy++GPTl+eefn3absVbkcj7Vb1m24XMnn3yyHX/88bbvvvumVCnb8SxB9/Of/7xtscUWttdee5k4vuMd77BVq1allJv0IdPx/Kc//cl23HFH22677ezlL3958F2TYB0PgX37298etOfLX/5y0uFYBgEIQAACEIDAOBJAwBpH+Bx64hDQDfeRRx45cSpMTTMioBsXCRY33HCD6WZ2LHvhhRfs9ttvD/bJV7TUjWKmN89j1WuirQ+5j1VvMRpL5BqrjHJbf+GFF9qnPvUpmzlz5nDV5EH12te+1jS+fvGLX9gjjzxiv/3tb2316tV2wAEHJN7Ey0vlkEMOsU033dT+/Oc/2913320f+MAH7MQTT7Srr756uOzwjThqubyl3va2twX7xG/cw231KkHhne98pyk87Ic//GEgWv3hD3+wqqqqQMiJe3xtu+22geDwrW99K1rMpHkvfvIyU/+MZtpuNK6j7RuumzVrlh166KEm0bCc7d///ncwLj7xiU+kVDOX8Xzqqafaj370I5NorXJ/+tOf2pIlS4IxnuSNmO14lnglLzEJsBK87r//fjv99NNN4/WUU05Jqb8+nHHGGfbjH//YFi9ePGIdCyAAAQhAAAIQGD8CCFjjx54jTyACeiJ+3333TaAaU9VMCch7RJ4HusEZyxReM23atLE2Yz0EEgksW7bM/vKXv5hClKL2zW9+MxA9br755sArT94nEq7++Mc/BmLpeeedF908eP+d73wn8Ly68sor7VWvelXgWaKbbnn2SQSIC6wKi/zMZz4TeGw98MADgfA1otDIgl/+8peBYKPXww8/PNheItuNN95oM2bMsC996UuRrTe8lVeXhDeJC5PR6urq7Kyzziqqh564NTU1BZ50ca+mcmMqbz8Jbc3NzSlVy3Y8y2vv0ksvDTzXjjrqKNtqq63szW9+cxDOqN9ejcG4ZTueFbYrj0Z5Pu6+++62zTbbBHm75IF1+eWXj8h5JWFYx0g6drwufIYABCAAAQhAoHQEELBKx5ojQQACZUhAN1+vf/3r7Ve/+tWYtZOAJe8VDAK5EJAnp0KX5s+fn7K7vE1OOOEEmzJlSspyhbcqnEneVVGT4HrdddfZMcccYw0NDdFVdtxxx9lLL71kf/3rX1OWq3zl1pK4IIFgLJOnoUK5FDoXNX1fPvrRjwaeiNHlei8hTSJPJmJwfN+J8FnC47p16+z73//+RKhuUesokVLijgSsuGU7nuXVtvHGGwfeVtGyJOTKyzApLDab8aycV0899VQQDhstX+81lvUduvPOO+OrgrbpO4tBAAIQgAAEIFA+BGrLpyrUJFMCCi/RrE/KMbHHHnvYG9/4RpMXSZIpl8v1118fuPnX1tYGT+mVV0I3GUkmD4B77rkn8EhRSIgSmba2tiZtGoQ/KbxErvgKmVCuGj01TVe2ytVFom4A9GRZoSy52PLly+3WW2+1hx56aDhfjG4K46b8LQqtUc6ilpaWoJ66sdJ7PYndYYcd4ruM+Pyf//zHHnvsMVOuGd00huEjYh73xFHIjby0dMxMjqEZyBTKoPKVyP51r3tdkJtjRCVGWaBcNHfddVeQJ0f5aeS1ofwe6Uz5d+69915TiIfGzv777x94U6TbPts+K0SbVBclz73tttuCtukmRmEfyl0Smo6jcB6JSfGn/9pG3xH9KafQWKYwtSOOOMIUBqOxOXXq1MRdlPRX4+HMM89MDNEKd8q2T8L99JrtmM12++ixsukrfefkWfPMM88E+ZD0/X3FK14RLW5c3mfThmzPhWGDCjnWdCMuD5OorVy50pQcPV2uLyV613k2ahr7+o7ofBs3eY7stNNOwfcnmmNNuayyMXnFKB9Rkm299dYmbzKJGFEBrbq62t761reabvolvGVqmfZNruNd5zt5vuk8Ls8bCdY6X2ZrCu2TF5vCzjRTnT5nY5n+dsl7ToK6cqElJfrP9hyT7W9TJm1S2LVMXnlRy2U8S2zVb6rCU+Om5fre6PdXnmmhZTOew3DXpPGs6yIxVjL6uEk8U2ijEuGn+37G9+EzBCAAAQhAAALFJYAHVnH5Frx05YiQW7tyM+gJ/Ac/+MHgCWVSolMJPNpWF3q6QLz22mtNsxkq6W98e+Xi0c2+BCtdLOpi/6STTgqewCsfRdy0vy7ulHNF20pQUtkShpS3Im5nn322HXjggXbBBRcE2+pY2jfbHEASfJQXRFN2Syz6wQ9+ENyMKDwgnndEuSv0dFWvenKuP+2vOuiiOxMvAbVNoTJqn26e9F5/uhGJmo6t8Bn9ZXKMa665JmjHueeeGwhxStYr7wX1byamGxLtq77UzZTEGnkFqAyFSMRNN4hioRtePR2X6BUm3pUok2TZ9lm+bQrroJtN3WBqPElwUyJdiZ2qd2hhe+JeJuF61V1jIxOTgKWxrJun3//+92l3kfeVbnrjHinhDtn2Sbhf9DXbMZvt9uGxsumrm266KWjzZz/72eA7p3EnofSTn/xkWNy4vGbThmzOhfHGFGqs6TsqETTuwSexWwKVvrtJJiEp7jGl848eFMydOzdplyDcTwJTPiYRTA9JkkzLJQRFxatwO32XNGbSzVgYbhe+ZtM3uYx3MdfDAXmm6XzxjW98I/CC03k6F/v4xz9u06dPNyWsz8ay+e3SOUnna52no5bLOSbb36bo8UZ7r+sPiUvxB1a5jmc9qEiycLm8qHI1TW4gcVXiWpJpPO+yyy4jVmkiBT0cI4xwBBoWQAACEIAABMaNAALWuKHP/sDydNKTbeUvkReNXiWwyINHeTmipqeVErf0xFFP6yUE6AZKrxJi4rPrSNRRWbqB13a6cP7nP/8ZhLRISNAFddQ+/OEPBx4AErd086UbA3lX6cZKHipRU7m6adByeXBoez3tlKdQNgl/leBY3mPyxpFXhPLD6MnoT37ykyB3RroZm5SgVdODq10SrfSqJ9ua8SieJyZab73XvtpeeWUU9qP3+pN3WtS+8IUvBDeTWjfWMcRV4pFEFm3/u9/9LvCQU2JnCXO64RrLdPOgGcSUB0dhQfKyEwsJeUpMq5vCqEl8UL3UvxoP8hJTEmg9eVYISHy68Gz7rBBtCusrNkpErTGisaIbeOXg0WxXuoGV6eZaN/VJIqREUXkoyvMuE9PYVuiWxNfRppDXTYyE13RjJts+Ga1u2Y7ZbLbPpq/kkSYBPPzOaVwoFEcirvLG/PznPx+tGUVbl00bsj0XxitdqLGmc59u9uNilLxKFKqX5EmousgLMu5hqvPsnDlz4lUd/qwE8UkPEoY3yOCNRB/9zigcMW6aRVHem0m28847B78XGidjWa59k814l3AlLxv9VmncyHtJv1fyuExKDj5WnSXaffGLXwweNow2Q2S0nFx/u6Jl6H0u55hsfpvixxvts8azhKG45TKeJSClE2PDyQ7yEbAkOMrjWPziJk9fTZaQbjzr+6/fWAwCEIAABCAAgfIggIBVHv2QUS3kmSIvG4VvhCYhRp4QupHUzWZoypmiG4jvfve7KSFlmmpd4lP86bMu6OWdoxvV0I1fTyTlhSVBJXqhrrL//ve/B3VR3orQ9EReSYR1ox+96dGU4roB001HaBIMJLpFw8LCdelev/3tbwc3bfJSij7510xZEpguuuiiEUKMyhIHCWj19fVB0bpRlKCjNkXble64mSyXyJLpMbSdwmwk1IQm5hLUxPDiiy8OF6d9FQt5OqgvoyYOegIe9cJSElx51Ul0UP+GphBIebFoFr7o9lqfbZ8Vok1hveR9dfTRRw+HxermU32rdkk0CU3tl4Alr4SoSUjVjWk2Apb21wxVEszi3olap9AjfQ+0TVzM1XpZNn2yYY/0/2Y7ZrPZPpu+0k2/zjsSRZW4W6Z+0M2/PCM0K914WDZtyPZcmNSeQow1nSPS3aQnHVPLdJ5V38pbNWoao+lCu7WdwmB1U56P6Xuom3o9CNEDA1lHR0cQUiUhWeeaJAvbmImAlmvfZDre9ZBHgr28paKheAoH0wMPnftyMQnZKiP+sCZdWbn+dsXLy+Uck81vU/x4o32WKBf29WjbRdcljWc9EJC3XrrxHIZ05zue9fun3zv9toXn8NATWdcD6QRhPbiKXs9E28N7CEAAAhCAAARKTwABq/TMcz7iG97whhGePypMuUY6OzsDb56wcHleKeQqdL8Pl+tVOVJ0URs15dDSjWrcdMMur69org95DOniPSknhLxYdEEqz6LQlAxYIlPc5NIvj6pMTTl4JGzEQxa0/7HHHhvUX8eKmy5co7kztD4UzuTlUwjL9BgSViSSxG9IVQfxOPjggwOxZLQ6SahU2F98JjPtI5FOQmSUvzzVxOzII48cUayeTCv/k/o0atn0WSHaFD22PCbipvGpsaWb0dAUiqUn99FlWicvIeUUyyTHmbYPParEXjdR8maLm0RZeZboaXx48xPdJts+ie6b9D7T8RTum+n22fZVKBTHw8FqamqC80JSX4V1yuRV30N5JKX7SwoDzbYN2Z4Lk+pdiLGmG355gmZqElfkkanvbXi+CvcNx2z4Of4aPoSIL8/ms84l+g1RneW9ou+UHlhItNR3TFyTTOcaJZ/P5KY/177JdLzrPCYBRCGvcZOHWejdE1831mfx/epXvxp4tCYl/47vn+tvV7ScXM8xmbKKHmus9/LY1TmhEOO5FGNZ7ZEYKw9bifESpRQOrnO+JkLQw6N0tmDBgozGcrr9WQ4BCEAAAhCAQGEJ1Ba2OEorJoF0NyUSpGRheJXe68ZAf0kmoSSeL0oihnJ7aIYqeXSFIpGEH3kLRU3iV7pE4bpZkCiimzWZntjrhjOXhLnRY6oMPYGN38iF2+iCVB4iYbLWcLlek+oa3rjEb8yj+2XzPtNjhFySttfxxGms2fDCfg77PV5P9aHCREMTE42FuIgXrpcoExVtsu2zQrQprIte041zibHyqAhNudB0AyUvrGhCceXfSUpuHe6X9CpRSjfsEoPFXzc1UVPIlEJpZUkCVrZ9Ei076X3S+BhtzGa6fbZ9teeeewai+WmnnRbk3dNYCU1T0edrCnkNzzVJZYXhtdF12bZBY19/SZZ0LkzarhBjTfXO5ob/wgsvDM7po91cJ9W1kMsUcq6wOz2A0AMLeXNKjJFHp5Jbp7NMvVZy7ZtMx7vygCmPkQTXQpvy8snLUyF68mBOd97K57crWudczzGZsooea6z3GgeyiTSelcvue9/7XvAbK0FaDysUPqgHdMr9GQ/tDRloLOvaIz5hQbieVwhAAAIQgAAESksAAau0vItyNN14KyQvKWmvLjSVI0rhT7oAVrhVUiJTPeWXN8tXvvKV4OZETynl4aPwrbipTIWUKG9LkunpbBiGFeatSOeen7R/0rLwgjkashjfTk9Kw5vb+Lr453Q3G/Ht8vmcdIywfmKrm+e4SXgK2cXXhZ/D0Jx0TCXkheFe2kfHTPLEC8vTDZ6873SjpYv6bPusEG0K6zLaq7w6osnzxVfi6o033hiErmpfjRPlK5GIl41J0NVYl8ehwsXEQDcuMoU5SbRV2JAsLv5qWbZ9on2ytaTxNFoZSdtn21c6t0i8e//73x9MGKBJChQKrFDXQphmcgu9vJLKU54d5SyKWrZtCPfN9FwYbh99LcRY0/lXXnyZmDwsFfamUMmkc57qEw+djZardUnnl+g2Y72XEKzfBYWny/s3NM3CJwFAvzfpchhK2Ai/E+F+o73m0zdhuUnjXd/jdOfJcL98XhUGrzB+fUeUpy/JCvXbFfJM1574eT+pLuGyJFbhukxewzZlKmCNNp7DuqQbz+HycLtM6pe0jc7resAmQTbMN6fQQYVlanzLqzCe21LlhGGS4i/BFYMABCAAAQhAYHwJjFQnxrc+HD1HAvJiiN9Ya6ZCeU5o1jS5z+siTU+iFfKQFDonzyt5oJx33nlBwuxzzjknSFarm/qo6UmkbsRG83IJZ9QKZxnM9wm4jikbzVtDN9u55jSJtq+Y78P6KYQwXVvGYhUyTRIXk+quY462bViPsG5h+WPVIzxWuF8+bQrLGu01aYzrpkR5sXRDJZFONyHK7aXZGXOxfffdN/Aq0A3pCSecEBShsBPN5pnO400bhcxG45xLfQq9Ty59pdxB8lT4xS9+EQgWymcnj02J3RKNS225tCHbc2FSm/IdaxLqxgqX0nH1AODDH/5wEDIb9wQM66WQ7qiXZbg8fFUZ0bDvcHk2rwoflIAVFa+0v75nmkBCdVQYXZIAqZD2TL8LheibdO3SWMm0HunKGG25vJvURwrTk6iXxKJQv13ldI5RSLcsfs2RxGqs8azfGZ2z041n7S/LZzwvWrQomEFVE52E4lVYV02eIk9PeRbqeilummhAFv5OxtfzGQIQgAAEIACB0hJAwCot76IdTTcMYbJTHUTJl/WkXDcYCg2Mm25Okkx5TpT4XV4WmmlJU3krBC0aLqKLV4XyaVa4sUwJn2UKS8vHwnLCi9mksnQMXQiXs4U3OEcddVTOF+ThBbj6PBMTu9CrKml7cdPTbXnxyULWmfZZIdqUVK/4MrU3rGO4TknpNe4VRqixqtkVDzrooJxvWuW1Ik8riVYSsCQ4SMzSk/rRLNs+Ga2sYq7Lta90k/m+970v8MqUR47CpiSE33zzzSUXsbJtQ67nwng/5DvW5Mkx2vcwPJ4eJEis0Hk4nYVhTenWjzarW7p9oss17pVwW7OcJpkeiCgkS94sSvYeN3n5apuxrFB9k+448ijNJBdXuv0zWa4w2GuvvTaYCCMpsX14Ps33t6uczjGhV5LG82jeveKXyXiWJ1c6z2ONZVl4zOBDlv8oxFNeo0l11W+f8qHp4UeSgKWxrPNf6JGb5aHZHAIQgAAEIACBAhOoLnB5FDcOBJTcVTcTUXf+G264IZh1KUm8yqSKcpVX6KGeLsvTIgzb0b5aF/08Wnny0NAFYngROtq2o62TF4jEhTB0IWlbrcs311ZSuYVcJnayeBL9YGGG/4SeQOluhvUkO9o/yu0R5k9JOoS4KYwwfMKcbZ8Vok1J9YovU/hg/CZCdVZydwlY8rbQTXemsw/Gyw8/y7vo3nvvDYRbzdap/CdJkxCE2+s12z6J7lvK9/n2lb6DErLEWSFkZ599dimrHxwr2zbkey4MG5jvWMtEwLryyisDTxB5JSmXYDrT90C52NKdD/X9z+eGX2KCworT1UH59PSXbhZX7Z8kFsTbU6i+iZcbftZ5Ld/fnrCsdK8KbdbDIoVTJs2UV6jfrnI6x4RhjElpC6KcshnPSV7hKisc4+Exo+Vn+l7jNN1YVhkKv0w3liVg6TcxU4/kTOvEdhCAAAQgAAEI5EYAASs3buOyVxhCED94mCMmvMDVel2MxW/2w/0keEVNT6jlvbJw4cLo4uD9Jz7xieD1/vvvH1639dZbB8m0w1Ce4RUJbxS+oRwuDz30UMJaGzWPS3QH3TxKiJGgkGSqu1z9k3JYJG2fy7IwF0cu+4b76OZbF8LKq5SriafKUK6nJNMT7zDhuNYrzEU3B+lEM9UlmtMo2z4rRJui7Ug3rjSGkm6KFbojQUXeQArXOfDAA6PFZf0+TFwuzyvNPqgZzMYKX8m2T7KuVIF2yLavNAugvDjjpnONEh/HZ4CMb1eMz9m2IZtz4Vj1zWesjSVgKcxJnq5nnnlmyqQESXXSGJXnoWYYjZu+6ypLXiW5msQCla/ciUkmQVwekdHzRridwsok5CR9V8NtwtdC9k1YZvRV4pEEkKQZdrVdIc7pKkfemvK0Ssq9V6jfrnI6x0jw0e9EuocoYpLNeJa3nnIZJoXY6uGE8m3Kmy5X0zjVWEvX3xrnSWNZx5OAJfYYBCAAAQhAAALlQQABqzz6IaNayMX9lltuGbGtntZvv/32KQmCdcF1zz33jMgroQs13ZDqQjGcTU1PkJX/4YILLhhRdii0RHPdKC+KblB+9KMfjdheF4jxm1p5tCjsKB7yppsKzQqUqR177LFBOUkzDarumiVLuYqKYbpg18W6Ep3nY7qZUa4oheYklaXk+OlCKcLjqgwlDBb/+AW/wlQ0zX2Ug3L36OZZyWrj9vjjjwc3wccff3zKqmz6rBBtih78jDPOGJFb5amnngryMCUlSlbIoEw3/sp9NdqT9uhxRnuv9issSB4iej+WZdsnY5VXrPXZ9pW+5+eee+4IIUPjTl5q8nAJTV6gmsEx6v0Xrivka7ZtyOZcOFY98xlrEt50Mxyed6PHkviuhPbKwRY+NIiuj7+X95MERJ37wxw94TaXXXZZ8H2P564K12f6qv2///3vJ+YV1PlL50QJaXELPWYyuekvZN/E66HPEhwltFxzzTUjVkucTfKYGrFhBgsUVq/zz6WXXprIqxC/XeV0jpFXtcZzOgEr2/Esr0797umBQdQklP7+9783TR6Rj0kgU19fffXVI4rRMXTeSvfgQ+ezTMTYEQWzAAIQgAAEIACBohBAwCoK1v8VesUVV5iSHksoiFq2y7WvylEunvPPPz+YmUsJSZWzQTfa8dwbCnmSx9Zxxx0X5AXStgpx0Ox373jHO4KqhElTlfxc5UjE0pTtmjpdIpFuhBQ+qJuU6I2KLlxPOumkoC5aL08gXRwq0bMuRPUXzZ/0sY99LLjJOvTQQwNPGXkCKSG0hIHoDXCUT9L7o48+OrhoPvjgg4OLWoWUaXajE088Maj7l7/85bRTmSeVl82yXXbZJXh6qyfsYjNaPpOxyhVjiXnKHSOvIV24a3woGbny7Gha77FMnhraR/maJFRKDJQXmm6U5DkhL6zQdHMlNipfCWvlyaTQj+uuuy5IxK/p4DWbX9Sy7bNCtCk8vuovkVTikcaKBDmJcBp3eo2bPB90sy0e+YYPhmWLq8a1PLribMJt4q/Z9El831J+zqavNA4kfupV5yyJE3/729+C75xExWiS8ba2tuBzKKppR7QAAEAASURBVHoXs03ZtCGbc+FYdc5nrIWCUhIfndfFU+fnW2+9NXhQoYcV0b+4F6tyEGpmNOXT00MDnZc0a+FFF11k/+///b+8Q540iYcECk3Wcfvttwfn+EceeSQIl5P4o3N/klis+us7rO/rWFbIvkk6lkQ2eaOKx4UXXhicI8RKD3GuuuoqmzlzZtJuOS2TyLLTTjslevkU6rernM4xEnOVAy3Jsh3PEjJ1raIxrX7RuJZXrc7Denim70U+pklnVL5+F/WwS981hZZKHHvjG98YXIekE441nnffffd8Ds++EIAABCAAAQgUkABJ3AsIM6ko3fDpxiIewpDtcpWtxOm6CNMFuRKsy+RWL2Hqgx/8YPA5/OcVr3hFcMGuG70wzEThdZqaXaKRclNI8NCTSZkujJUkVjdA0eTBEgwk2ujpb9QkiOgGRa9h2ISedGtKcd10hYlrtY+eXiq5tuoYChC6cVDyW4VmRcMTo8eIv1fyZiVjlVgnb4XQk0EhjWpj1Osovm++n1W2hD+Jh/rTE3093c/FdEGuC3+FnegCPfSiUtiOmEiQG8vkcacLawkL0SfHqqdYxG/MJCrqZlMX8KHnnIQtLVefxy3bPitEm8I6aDxpDGqsh15qe++9t2nigfg4DPdRXygEZbSZMcNtM3lVOIlm71TYanQsj7Zvtn0yWlnFXJdNX2lMSkDU+SEqkkvUkvdk+H1WfcPvYykmUsimDdmeC8din+tY07lO41gznkk0jppu1vXAQaHc6UwzxOrmPjSdz9U3EkdCcUzH0O/Bh32GwHxN5wA9lJCoEBWGxV6eLNG+jx5LnsI6r+nByFhW6L5JOl6Yp+2ss84KhCzlcVM/6IFNoc4XOq68kiSMJf0uFOq3q5zOMRJ+LrnkkuABTDzEOpfxLE9PhcaffvrpwXlf/aTrCY35QpxTNHOmcmRKmJW3nEzHkCejfnP0exg3pWfQw574TMzx7fgMAQhAAAIQgEDpCFR5PqSh0h2uMo+knD5JF/PZLo/S09NDee4ov5EuwtKZPEj0tFFPonXjMZYpf4kEN+XJ0oyEmVw4ygVfHli6uE5qZ/SYeqIvzy/VWxf8uZra9dhjjwVinEIgS2EKj1QeDTGSaJZOTMmmLgq7kueQBIF0OcvGKk88FXqoi/P4jUTSvnq6HfaXRMexLNs+K0SbVCdxlheUns5r/I5mupGSMCdvtHKwbPtkvOqcTV/Jw+qZZ54JhFCNtXhSY4nL8qjR9zKTc02h2pxpG7I9F6arXz5j7bzzzgvGqTyZCmn6PdBDEp1Xi2ESknXu00Qho/WtRDg92Lj++uuDvEWZ1qVQfTPa8VQ35WVSXix5iI2XFeq3a7zPMbp+EUs9AIk/QMuHrc776ifl2StGP+l3PPSi1vclnNU0qc7y1lLeTc1Ki0EAAhCAAAQgUB4EELDKox+oBQQgkAcBeacozPVLX/pSHqWwaz4E5N0nr0CF9U5my2esacIMhSPddNNNKWHZk4WXQqLlRar8aNjkJ3DKKacEYd5KCTAZTRN4KF2ChHkMAhCAAAQgAIHyIJDedac86kctIAABCIxK4IEHHgiEEwkL2PgQUHJyhZUp6fdktnzHmsK3FRqrZN+T0RQGVkhvnMnIaDK1Sbkz77rrrsA7bzK1S22RGC/v9Wj47GRrI+2BAAQgAAEITEQCeGBNxF6jzhCAQJAsWfnT5AWgMEPlyMLGj4DCWJUzbDKawo4KNdYUvrvHHnsEkygoz9pkMeX2kgekJgHRLIlYZRBQMn9NMKCZhieLKTxSuSWVk075NjEIQAACEIAABMqHAAJW+fQFNYEABLIgoIT1yiWkZNOaOCDXHGJZHJJNK5RAoceaJrVQYnx5eWSSt67csSuPn5J6a0ZcJd7GKoeAcnodfvjhpnA7zXI7GUxhg/Iq1XjGIAABCEAAAhAoLwJjZ3Aur/pSGwhAAAIBASXE1sx3r33ta0edyABcEMiXQKHHmsSejo6OEYnw863neO2vmVQvv/xyxKvx6oBxPK6SoEvoCWc7HseqFOzQ8ow88sgjC1YeBUEAAhCAAAQgUDgCeGAVjiUlQQACEIAABCAAAQhAAAIQgAAEIAABCBSBAEnciwCVIiEAAQhAAAIQgAAEIAABCEAAAhCAAAQKRwABq3AsKQkCEIAABCAAAQhAAAIQgAAEIAABCECgCAQQsIoAlSIhAAEIQAACEIAABCAAAQhAAAIQgAAECkcAAatwLCkJAhCAAAQgAAEIQAACEIAABCAAAQhAoAgEELCKAJUiIQABCEAAAhCAAAQgAAEIQAACEIAABApHAAGrcCwpCQIQgAAEIAABCEAAAhCAAAQgAAEIQKAIBBCwigCVIiEAAQhAAAIQgAAEIAABCEAAAhCAAAQKRwABq3AsKQkCEIAABCAAAQhAAAIQgAAEIAABCECgCAQQsIoAlSIhAAEIQAACEIAABCAAAQhAAAIQgAAECkcAAatwLCkJAhCAAAQgAAEIQAACEIAABCAAAQhAoAgEELCKAJUiIQABCEAAAhCAAAQgAAEIQAACEIAABApHAAGrcCwpCQIQgAAEIAABCEAAAhCAAAQgAAEIQKAIBBCwigCVIiEAAQhAAAIQgAAEIAABCEAAAhCAAAQKRwABq3AsKQkCEIAABCAAAQhAAAIQgAAEIAABCECgCAQQsIoAlSIhAAEIQAACEIAABCAAAQhAAAIQgAAECkcAAatwLCkJAhCAAAQgAAEIQAACEIAABCAAAQhAoAgEELCKAJUiIQABCEAAAhCAAAQgAAEIQAACEIAABApHAAGrcCwpCQIQgAAEIAABCEAAAhCAAAQgAAEIQKAIBBCwigCVIiEAAQhAAAIQgAAEIAABCEAAAhCAAAQKRwABq3AsKQkCEIAABCAAAQhAAAIQgAAEIAABCECgCAQQsIoAlSIhAAEIQAACEIAABCAAAQhAAAIQgAAECkcAAatwLCkJAhCAAAQgAAEIQAACEIAABCAAAQhAoAgEELCKAJUiIQABCEAAAhCAAAQgAAEIQAACEIAABApHAAGrcCwpCQIQgAAEIAABCEAAAhCAAAQgAAEIQKAIBBCwigCVIiEAAQhAAAIQgAAEIAABCEAAAhCAAAQKRwABq3AsKQkCEIAABCAAAQhAAAIQgAAEIAABCECgCAQQsIoAdTIUuWLFisnQjJzbQPvp/5wHzwTfcWhoyCp5/A8ODtrKlSsneC/mXv2BgQFbtWpV7gVM8D37+/ttzZo1E7wVuVe/r6/P2traci9ggu/Z29tra9euneCtyL36PT09tm7dutwLmOB7dnd3W3t7+wRvRe7V7+zstPXr1+dewATfs6Ojw/RXqaa+1xioVNN3v6urq1KbP6HajYA1obqrdJWt5AsYUab9lXsBS/9bRV/AS8Cr5At4CXiVfAFf6e2XgFnJ/S8Bs5Jv4CRgVvINnARMiViVaup/iZiVaur/Sm6/2i4GlWr67us7gJU/AQSs8u8jaggBCEAAAhCAAAQgAAEIQAACEIAABCqaAAJWRXc/jYcABCAAAQhAAAIQgAAEIAABCEAAAuVPAAGr/PuIGkIAAhCAAAQgAAEIQAACEIAABCAAgYomgIBV0d1P4yEAAQhAAAIQgAAEIAABCEAAAhCAQPkTQMAq/z6ihhCAAAQgAAEIQAACEIAABCAAAQhAoKIJIGBVdPfTeAhAAAIQgAAEIAABCEAAAhCAAAQgUP4EELDKv4+oIQQgAAEIQAACEIAABCAAAQhAAAIQqGgCCFgV3f00HgIQgAAEIAABCEAAAhCAAAQgAAEIlD8BBKzy7yNqCAEIQAACEIAABCAAAQhAAAIQgAAEKpoAAlZFdz+NhwAEIAABCEAAAhCAAAQgAAEIQAAC5U8AAav8+4gaQgACEIAABCAAAQhAAAIQgAAEIACBiiaAgFXR3U/jIQABCEAAAhCAAAQgAAEIQAACEIBA+RNAwCr/PqKGEIAABCAAAQhAAAIQgAAEIAABCECgogkgYFV099N4CEAAAhCAAAQgAAEIQAACEIAABCBQ/gQQsMq/j6ghBCAAAQhAAAIQgAAEIAABCEAAAhCoaAIIWBXd/TQeAhCAAAQgAAEIQAACEIAABCAAAQiUPwEErPLvI2oIAQhAAAIQgAAEIAABCEAAAhCAAAQqmgACVkV3P42HAAQgAAEIQAACEIAABCAAAQhAAALlTwABq/z7iBpCAAIQgAAEIAABCEAAAhCAAAQgAIGKJoCAVdHdT+MhAAEIQAACEIAABCAAAQhAAAIQgED5E0DAKv8+ooYQgAAEIAABCEAAAhCAAAQgAAEIQKCiCVStW7duqKIJFKnxfX19tmzZsiKVXvxiOzo6rKWlpfgHKtMj0H76v1LH/9DQkHV2dlbs97/S2z84OGjd3d3W3Nxcpmfn4lZL7e/p6bGmpqbiHqhMSx8YGLDe3t6KbX9/f7/pr7GxsUx7qLjVqvT269pd54CGhobigi7T0iu9/Tr3yerr68u0h4pbLbW/qqrK6urqinugMi1dv/3V1dUTtv0at3Pnzi1TuoWtVm1hi6O0kEBtba3NmTMn/DjhXnUDM5Hrny9w2k//V+r4l4CzePHiiv3+6+ZlyZIlFdt+3cCuWLGiYtuvG7hVq1ZVbPt1Ab927dqKbb9++9vb2yu2/Xp40dXVZbNmzcr3MmpC7q+Hl/oOzJw5c0LWP99Ka+xLxJ4+fXq+RU3I/d2pw3QNNG3atAlZ/3wr3dbWZjU1NTZlypR8i5qQ+69ZsyYQr1pbWydk/SW+VYohYBWpp6VgT+QnOBO9/vl2K+2f2OOX/s+dgC7eKnn86+JdFwET+fyde+9b0PZKbr/Gvi7gK7X/9f2v5P7X97+S+18CrrwwKnX8S7zSGKjU9kvAreTffzkf6BxYqf2v9lfy+U9tl/dZpfZ/PteOpd63cqS6UpPleBCAAAQgAAEIQAACEIAABCAAAQhAAAIFIYCAVRCMFAIBCEAAAhCAAAQgAAEIQAACEIAABCBQLAIIWMUiS7kQgAAEIAABCEAAAhCAAAQgAAEIQAACBSGAgFUQjBQCAQhAAAIQgAAEIAABCEAAAhCAAAQgUCwCCFjFIku5EIAABCAAAQhAAAIQgAAEIAABCEAAAgUhwCyEBcFIIRCAAAQgAAEIjEVAMzz19A2Z+f9V/gitv8+sxq9EGuqqffarsfZmPQQgAAEIQAACEIBAJRNAwKrk3qftEIAABCAAgRIRGBgcsrXrB2zRs9122a+X2qq1/TZvRpW944DZttOWrbbp/HqrrUHFKlF3cBgIQAACEIAABCAw4QggYE24LqPCEIAABCAAgYlHYH3ngN3+73Y7/6cvWXfPQOCFtWKN2YvLltsJ76uypkaz+bMaJl7DqDEEIAABCEAAAhCAQEkIkAOrJJg5CAQgAAEIQKCyCfT0DtlVNyy3nt4BGxo0c4csG/TXtvV9dsPfV9ryVS5qYRCAAAQgAAEIQAACEEhDAAErDRgWQwACEIAABCBQOAL9/YPW3tlvngZLKbCGTZ+XrxqyPqlZGAQgAAEIQAACEIAABNIQQMBKA4bFEIAABCAAAQgUjkBrc60113uy9niRvqDFwwdnTKmLr+EzBCAAAQhAAAIQgAAEhgkgYA2j4A0EIAABCEAAAsUiMLW1xg553Qyrra+xar/6kJClmQcbGqrtsINm2fyZCFjFYk+5EIAABCAAAQhAYDIQIIn7ZOhF2gABCEAAAhCYAAQ+dMhcG6qtthtuXW2dXQM2w0WrY94yx3bcqsUaG0b4Zk2AFlFFCEAAAhCAAAQgAIFSEUDAKhVpjgMBCEAAAhCocAI1NdV2zNvm2mH7zfTk7Z7M3RNgNXpY4ewZde6VhYCVz/Do7x+ybk+UP+hM62urXBCstt6+QevqHrTO3kFraaq1qc3/c7xXyjF5wmEQgAAEIAABCEBgohBAwJooPUU9IQABCEAAApOEwPSptTZtSm0gttQgXOXUqxKgevskApr1+wSObe39NuBTOw4NVVl11ZDVuKfb6rZeu/W+dlvT1mdVNWabzGmyV+7aavX1Vaak+i1NNZ57rMaa/VXhnIOaGhKDAAQgAAEIQAACZUoAAatMO4ZqQQACEIAABCYzAQkmNfoHy5pAd8+gPfZctz3yVIetbu+zrt4qmzO1xqa6KNjSVB14YC1e0WN3/afd1nb02/rOAffOqrI7B9fZzXevtlfs0mq7bdsSbLtsVbU1efhmg3vCDfT3WZ+X3Tqt3+rci0um5YiMWXcRO0AAAhCAAAQgUAQCCFhFgEqREIAABCAAAQhAoBgEevvM/nj7Krt74Xpr73JPq6pBT4hfZS95Dvy50+ttamute2KZPb+s25as6rfO7j7r8O0UYijr6e2x+x8esq6eAdt1m2Zr9dDCzu4qm+4CWFdnvz39Yq899EKbDXg4ory2VN5uWzfb5hvVu5cWgmMx+pQyIQABCEAAAhDIjAACVmac2AoCEIAABCAAAQiMO4E/37XGfnvrGlu9zsMC/b+6Os935SGBUzy/VX39gLl2ZQ2+bOUaF646e63DhaiBfg8z9L/QVnm44RPPdplHGtqs6XXuuVXn4YNmi1fIq6vXc2a12UvLXClzvWr6tDq7/5F2e8u+M2zvXaaERfAKAQhAAAIQgAAESk4AAavkyDkgBCAAAQhAAAIQyJ7Av5/osJ/+fpm9tLLXPNovsGq/kmttdCGrut76e13A8hDCtev73dNqyHp8mwH/63XxKvSdUs6sTg8TfGFpj4cfDtrMadVBKGFrQ517ZfXasjUDtsLFLy/KzLdduW7AVvnngYEh23azZpvhnloYBCAAAQhAAAIQGA8CzD8zHtQ5JgQgAAEIQAACEMiCwKq1/R46uMbFpV7r7XWPKveY0l+fv1/fLbFqyGprN4hLq9cNemL3QautVlL3DQfRi2YdrHHBa9DFKQlU3T0KMRwMxK6XVve4MDZgq9cOenJ4X9/jf152V9egJ4jvs0ee7fSwxbVZ1JhNIQABCEAAAhCAQGEJIGAVlielQQACEIAABCAAgYITWL22z154qScQl+QZFZreKkRwis8mOHeWe2G5+KTE7hsErWprdu+sZk/E3lBv1tjgApZf+TV4gvZGJW6vqwlyY3W7t1Zvz5Ct73IPLgljLm6p3KBs/ywPrs7OQbv34U5bt16uWRgEIAABCEAAAhAoPQFCCEvPnCNCAAIQgAAEIACBrAj0eQhfr4tJ1R4LWKXHj/4+tBkuXr1s+xZbMLvBbr6328MGPbm7tnGhqt7zY9W7Z1aL58hSWOGQ7z/oZfX7Nv2DnqjdP+t9vQtazf3V1ulJ4avd80qJ4EMb0nv35mrr6LXHnu+y3bdttvo6noGGfHiFAAQgAAEIQKA0BBCwSsOZo0AAAhCAAAQgAIGcCUxtrrE502rthYZqG5Ci5O5R8pCq8yu5jefW2ibz663PvahqXGhq9m17fJbBwcEq0+SDje6F9fKdWq3bwwHlmbWuo99zYLlK1TcQCFh1PoNhi5fb4H917b7MVS7lx5I3loQwJYqfObXWmj2/1goPNWzvaPDk7whYOXcmO0IAAhCAAAQgkBMBBKycsLETBCAAAQhAAAIQKB2Bjec22K7btNjSFX22dE2356eS91SVzXJhaZtNm21aS52tcc+q5gbPg+Wi1YAnyHKJynShN7W51nbYrMU2mldvC5/sDJK817r+tGrtgIcHDtrUlhqbN6vOPbQGbVl1ny1v2xBKqNZJK5viwlWr/225oMn6XBHr9+NgEIAABCAAAQhAoNQEELBKTZzjQQACEIAABCAAgSwJKNTvwFdOtY6ufntuSZ11+qviCRs8PHDOjHpPyD7gCdkHXMhyD62BAZs53T2yPFN7n3thbbN5s223RaM1uYfVntu12BKfxXD2tBr33qqy2poq63bhSjMW9nv29oVPdnhZQ0GookcYBpGKyo21YM6UQOhqbpQnFjMRZtl9bA4BCEAAAhCAQAEIIGAVACJFQAACEIAABCAAgWITmDmtzt70mhn29OJue+bFblvjCdXr/UpuTXtvsKzTRaz2Dp990EMCp7rINMu3b26ssde+bGogXql+M6bW+F+zzyxYb10+C6HCDNes67eV6/rseRe2lq1WCOGG0EF5cilQULMdrvX18tSaP7vOWrxMDAIQgAAEIAABCJSaAAJWqYlzPAhAAAIQgAAEIJADAXlMTWuttS0WNNjQkCdeX9Hr4YAeUriqzxY93WPdPn2gcqu3elL3KY3Vtv1Wrbb1xg2e36rK5E2lBPAy5bXaapMGW7mm2hYvd9crtykuSi1f3R+EDOqzF78hYbzvo90a3fNqiocizveZDrvcY6vFy8cgAAEIQAACEIBAKQkgYJWSNseCAAQgAAEIQAACeRCY1lpjXR4qqNxUta5I9Xmiq/883mHr2oeGJyZc58nap7gH1nRP5j6lacOlXpW7U7m+ZdW+jxK21/hrqwtSG831z02etL3NvbHkeeXOVfK4UmESriR6Nboqtvv2zUGo4aJnu4LE8dNdSNt+i+YgCXwezWFXCEAAAhCAAAQgkDEBHp9ljIoNIQABCEAAAhCAwPgTaPVQvikttTbVRaSnFvdaZ+f/xCvVrtdFrcef67bnlnYHgpVmE1y5tt/Wechhp4tbA/9Nwt7lIYfyxmp1kaurb9BmTqn1HFouYnkZyrnlelbwZsetGq3ON+zoVsjhoPX5TIZLVvXaE891BZ5a2gyDAAQgAAEIQAACxSaAB1axCVM+BCAAAQhAYBIQeN5zLj3wRIdV+/R1u+/YYJvNbp4ErZqYTWh2D6pZ02rdI8oFKRehXHsaYf0eGbh4aV8gVvW5YNXv7lf1dTUuSg3YkhUuQvV7oqz/Wo+7XMmjav7sGp/psMFza9XYS8t7rUqhhhs32wF7TbEOn/Wwp8esoc5LkCLmtnR1r82YXuszIda5UPbfwniBAAQgAAEIQAACRSKAgFUksBQLAQhAAAIQmCwEbv3Xerv57jUuhvjMd253PVhj73z9dNtrp6mTpYkTqh0KA9xy48ZgNsDN5jbao091Wu+GrhluR42u8FxUWrl2g4jV415TzfVDNsVDEOt83WpP3D7DPa6G/L8+V8Aa6qts3sx6D0XstM1dtNphs+Ygl9Ymc+uDMqv8ANU+Y2Ffv89Q+F+xSl5a/f65s3vQwxFRsIbh8wYCEIAABCAAgaIQQMAqClYKhQAEIAABCEwOAl1d/XbLvf8Tr9SqgYEB+9Od7QhY49jFEo/mz6qz9x480+5+eL0LUr1BXiwlX9e6KR5mKEepVW191uPeU/Ka6vOww2qPD6z2Daa79lhXX20DQcIr5b6qspnu1bXF/DqbPrPZ1ncM2KwZdcHMg0FCdxfNulyoUtmyoAz32qoNRC15ZCFgBWD4BwIQgAAEIACBohFAwCoaWgqGAAQgAAEITHwCz/lMd319Mfceb9b6jl5bubrbZs9snPiNnMAt2G7LZjvyLXPtZzcus44ujxscqrJm94aa6+LTgtkNgeC03oUnqVtTW4KsVp6s3XNmKaF71ZBNnVobJG/v8ZkFm+vrrb+3OggLbPUE8FNdBJMp2bvev7isz0MW+63ew0hn+H6zptcF3lcStSRkNfnMhHrFIAABCEAAAhCAQDEIIGAVgyplQgACEIAABCYJgdlTm9w7p3Y4fDBsVl1tLeJVCGOcX9/3xhm2xYI6+8M/1gSJ2hvds2qrTettk3m1gfeVMq0r+rPOhacqV5sULljj7zW7oGymi1Hr3OOq3UWtaa3VLlbVugBW42GJHi7oelSjb984pd622KjRlq3u87DBQS+rKjjWgGtj8vbqdgGsx0MRlUur1tdhEIAABCAAAQhAoNAEELAKTZTyIAABCEAAApOIwOzpVbbtZk226Jn2lFbtuV1Tymc+jB+Bbg8R3HazRjv28HmmmQVXrun3kMGhIAxwyMWrpoYaa53mIpTPNjjNc2BJlJra4t5S/xWa5EElUavbrwqbPOXVpvMbg9kJk1o0x5O2d/Z42Z0DQU6sFvf2Cr2uFGrY2eOeXrUbPLeS9mcZBCAAAQhAAAIQyJUAAlau5NgPAhCAAAQgUCEEPnzIbLvhthp78oUeq3KVYvftptjr926tkNaXdzMlUHW4mCSrd4+qbheQGn2WwhXuKTXgCdqbPayv0QWslib3rHLvKglVUzwkUF5Toa3r6A88tRQpqmTwbZ7gfdqUmmFhKtxOr0og39pUFYhgNQlprwZ8xkOZ6iGvLM/IFSSKt8Eq/3/IZnri+DCPVrAh/0AAAhCAAAQgAIEMCSBgZQiKzSAAAQhAAAKVTOCQ/WZUcvPLtu0K4ZNk1OuzDLat77OObrN2F6R6Pcyvvs7zYbmApb/A88qnD1T+qqnuhdXw3/DBfhecutyjKhr0pxxZ6zs9HNBFLAlkCjsMLZh10MWp9vUDvt+gC2HVgagVrpc3lmYl7OjaIKq1tffZrfeus0ef7bIhX7TJvAZ7/1tm20Zz6sJdeIUABCAAAQhAAAIZEUDAyggTG0EAAhCAAAQgAIHyIyAvKIUNPrO4J3ht7xi09S4eyQtrzsz6IHxQOawURqhlErNC8Uri1/2PdtrdC9vcW8psroeL7riFvKbMc2J5GKKLWxKzlO9KSd2llK1Y0+fH6rZ2P0aHC1Uqa/ftWvzVvbK8Li2+3Zq1G5L+KzH89be12W33tFlXX18gkj3+fI098UK3nfnRjWze7PrhMMbyI0uNIAABCEAAAhAoNwIIWOXWI9QHAhCAAAQgAAEIZEhA3lHtHfKi2uARtdrFo74BDx4cqrWurkEP9atyT6yaYKbBGveOam78X+jgvx5Zb7/7+0rr/K+31ItLB+zpl8w+PGfQBj2HlsQreWgp31Wbe1w1uQD2+PNdtrKtLwgRlHdWt4cULnyyw/Z72TRr9vXKvaXQQa17+qUeu+uBNmvv7lMe+UDAGvTpD59f2mk33bXKDn7NbJvjMxnKUwyDAAQgAAEIQAACYxFAwBqLEOshAAEIQAACEIBAGROQyKT8VxKv5JFVLVcod5eSV9Ws6RKs5EFVbTMi+af6fIbBOx9cF3hZKam7607B3/KV/fawC1Lyqmpr77c1/ifPqyb33Fq5xlx86rF2F7QUIiiRS6JVQ32NrfXtaqpqA++vzu4BW7NuwFas6g0SvgcxjkGNzLyqVuUC23PL+gMRTOLYzGmT/3J00B3bxFNeby2eQywqJJbx0KJqEIAABCAAgbIiMPmvGMoKN5WBAAQgAAEIQAAChSVQ7+mkFO7XWO/5qJSDyj2vZNWeNF1a1owpdTZrWmrOqSDPlYcJyupc/FJuK3Ota9DL6XHhS7MJrvRwQQlMQ/5fVXuVi2NVgQizvrN/OO9Vp3tbSbxa5zmzamp8ZkIXw1qaajzUsD8IPVTZXT3BYTb8o8N4pebNqA+8suSxNdltVVuvXf/3Nlu1tjcIudx0fr3ts+sUWzC7YbI3nfZBAAIQgAAECkoAAaugOCkMAhCAAAQgAAEIlJbA/Fl1tnhZr+eTMvPc7e7dU+VeUbXufVVnLY21NmfG/8IGw5opd9W0qfXuMdUVLKr1PFc25Pv58tkzXNByIUteVhKvZEMuNA25GDbky6U5ycdL1usJ4Ne0D9rTL3ZZk+fakjdV639nOezpq7GX7dhkdz04YL297rHl++k481y42WW7Zhe4JGZN7vBBhXb++IaVLgZuUPE0i2N7x4DP5lltb9q3LhAdN5DkXwhAAAIQgAAExiKAgDUWIdZDAAIQgAAEIACBMiYwY2qd7bBVs72wpNc9ofpMMwEqt1SDC0oSt+p85sEkO3if6fazP3s+q+4Nnli1tTW2YH6V7bZti720si/w3vKUVYGwpeTvCn9bMLfB6tvc48q9t/p85kPl1Zo9zZO+e46sTtfJVq7147sm1e3rGt01bN/dp7mIVm2PPNXlidw9Ubx7Xh3wihbbaFZ94KkVJIdPqtwkWKa8X/9e1Gmr2/7ngqawy04XtZ5f1u2ecmI0CRpKEyAAAQhAAAIlIoCAVSLQHAYCEIAABCAAAQgUi8BGPqNfIDINNGxIwO5i05SWWp8VMFm8Uj2237zRPvr2+XbXwnXW5aGA82cM2TYLPMzNc1pNbRl0r6k6W766zzyle5DLSjMZKjxQApmnaXdhpj8Qpzq65VnVb2s7zJ58rss9rYas1berq+2yzTdussPeMMfe+hoPMfR8WTNafLnvL08shTwGnl/FgpKm3F4X1555qcvzdPXbZvMabaO5xVGRlGdsfVd/kLw+WhWJWEp0Px5tj9aD9xCAAAQgAIGJRgABa6L1GPWFAAQgAAEIQAACMQIbcl3VWI+LM/L8CUWi2GYjPm4yr87eNW9WsLy7u9tWrVrlsw1WBaGHSwf7guTvvf1VnmPLwxPdUeulVf32jIcLdvf0b8iX1V9j01vrbHpLnQtCvbZqXV+QSF6qzTQX0Na29wXeWXvs0DLi2OOxoN09x667dXUgYHV4rrAO94JaMLvWDth7hotZnuurwJWaOc2FxeaaDbnEvF9kCqWcNsVnhnSXti4X/+Qpp9BCDAIQgAAEIACB0QkgYI3Oh7UQgAAEIAABCEBgQhCoqpJX0wYhpNuTsGu2QGkm9e7tJM8pX52RqZyprZ4Ly8MQ2zuqgtkCe92b6MnnOu25xT0uZA146OCG2Q47OuVtNeihiwO2yhO36xjKoyUVrcqP3uP1WLrKYwfLwCQc3flguz2zpNuFq0EX3Ny7zJc9u6Tf7n14nXub1drGM/utsbXfprjoJE8zmTypJAx6ozZ4jf13udaJc5f/qeFiL68yeaBpFsYHn+iwR5/q8H5QiKW2UZ58T6zvoZrLV/fbD361xGbNbLBX79ri3nDNeGQJKAYBCEAAAhAYhQAC1ihwWAUBCEAAAhCAAAQmGgEJKsHsga65KBl7r+tHyl81rXVkMvf0batyYara/+o9H1avdXQPeLL2fvfE8oJciZHDUL0LNl0uBOl4U9zBqrpqMFguAWfIE8JrYsM+P36j58AqB+v3DPeLV6gNQ0GdJV7JhrxJyvllVe5dtrjXnl3RFoRf7rFdi4dS1rgAtYGnZkzUrIoz3XtKubvanMfa9f1BOKDKqa2pDvKCPbu0y267b50tcW7qh25n1zdYbY11HnrpXDp7uk0amPjOm9Ntbe65JuFr602bVAwGAQhAAAIQgEAaAghYacCwGAIQgAAEIAABCExEAp3dg4Go0uniUjiLYLfPFtjigknoVTRWu9xJKJghcI0LNJ0+G6FCBjvlbeQhd/1DA1ZfV+N/LnK5kLPRnEYPw1NS9mp78PGOQDQLPL78INUe27jD5uUhzCjhfF2Nq1b+v/JQhSZhqt+FuSUr3WtMbavvC0S7xSt6ba8dml2Eq7Ien0WxU0KUi2ADA5rBscaa/Cq6zvOFicPqtQM+E2OHLVvVY88t7XUPq55AvOr17fs8/FKzOHZ2yyMtOHwQZtnh/dTmnlpr1/bbxnMaELDCDuEVAhCAAAQgkIYAAlYaMCyGAAQgAAEIQAACE5GARBaJVz2eKFzeUfKIam4Ycg8sT6LuMxZmas0uSK1e6zMNurvVWp9lUGV1e/hgv4fT9XpSLNewbIGLV1tv+l8Ba3W158+q8ZDBXt9m0JPK19qOWzbZvFnlcbmpduy+fas9t6TPwypdlHM+slpvSGNdnX/usQYPt+wKPKY80bq/3npfv+24RVMg/EkYfOI5F6he6rbVLuxJBNtsfqNtMb/BGtzL7MXlPTbk3l3tHT6zowuG6gflDXP8/qowxP+RlyOb/gZ9+6WeKP8f96+1rTZutL13aQlEv/9tyTsIQAACEIAABEIC5XFFEdZmEr0O+IXdunXrJmyLBgc9N8SaNRO2/vlWnPZXdv/r+1up43/I77Qquf367vf7HWel9r/6vq+vr2Lbr77v7e2d8O1v89kBl6zsD5Kn9/3X00jiTf9AvW0xrzZtwnD1fbT93b2Bq5J1ehjcc0s63XtIfFzAcjGmxb2OaqqqbfvNfDbDTV3Mqu6y/t5+a63zpOgzq13AqnJhxpO6u2fS40/32NyZmQtn+f6Gj7b/Jp6v/jW719jdC6vMq+upq4Zs87nVLl656KbcXtUDPnNgt4tOLtx5XislWV+yvM8afQbF55f22wOPd9riZQOB+CRh6oWlPfa0C3Rbb+wCmItRCtLUvoODHqro5QcilcSq/zl8pVTPNS7vF99noM/uX7TKRcFuF8Rq0vZRys4F/tDj7mH6q9TznyYxqOTzX1dXV/D7L6/JSrTOzs7gu6scgJVo6n+1XeevSjR9/3UNoOugiWg1nphy6tSpE7HqWdcZAStrZJnvMJFPAKr7RK5/5r2UvCXtr+z+16io1PEftjt8Tf6GTN6lYbvD18nb0uSWhe0OX5O3mrxLw3aHrxO1pbV+/9nl3kLKPyXT7ZhHulmPL1MYYEuanFRhu8PXahd3Ojx8cK3neupz4aq+dsjmTK9WBJ7NaK21BS78bLGgNkjWPuC5nRSeWOfbdLrwdcfCLuvo7A+8txY9U2N7bNtg++xaHqGEu21db/rTfdpqD+FTjqsXPOzvief9fYef/13006yAynfV3Ki06+bC3ZDnz+qx5SsVRrgBqnJn6V5nmSevn+HebVOn6sa/ymZMUeL7WhfI/Gaod8O2nh7LqlzDc53KfLdh88ME+cPm+UyIOoa8u1oaGoJk+E3eT8qTVSoL+z18LdVxy+U4YbvD13KpV6nqEbY7fC3VccvlOGG7w9dyqVep6qF2V7KApfaHf6ViznFyI4CAlRu3MfeSCjpz5swxtyvXDfT0bSLXP1+utJ/+r9Txrx/vtWvXVuz3X0/e1q9fX7Htl/eBnsJW6viX95H+Jnr7axvc+8oThQeJ3P0HsdZVEjlVTHPRacqUeps+JfnyT0+gu7r7rLZhWnAh39xSbW1dvdY35KqOyy5NjUpivqEsd1a0/sF66+prsqruGuty8UWzHVa7gHXfY2tdvKkJZj6s8+TkskefH7KX7dxq8z1XVjnZLPfIku2ynTfojlX2yJOrrbGpyWdSrAl4bTSvzsP8PAeWJ7IaGOrzNkuRcnNVS8KWO6H5ehemqmpt07nN1uPnkFqHPaXVzyUujK1c0+vbuUeWkre7GKg+UV4sJdZXIvcZU2ttOw9RnOahnQ0eftnc0mQ9g3W+3vNjdZnN7K+zLTdpdHEwOGpR/9G5r6OjY8KP/1whKXJCHmgT/fufa/v12y8PlEptfyjeVGr7df2n+9fp06fnOoQm9H66/mvwhweV4sU0kTsr+QpmIreIukMAAhCAAAQgAIEKJqAE69N9pjyF+ilBuSJimjwETmFwCiVMZz0uQrV3enig54aSQHXHA2126/3rbenybi9nwGcTrLXZ02oDEUsiTdP8ai/bcz8t6w2WqfwmF6zWeX6oYMa+pv+pLpoN8cUVfWUnYIUs6t3b6qC9W2xKo3uO9bZYg7djzgwXlMTMb+oUarjouW579Mku+//svQmUZOd5nvfVvq+9r7OvGKwDgACxcJMArqAl65CUKUqyZNHHdmIdnShxnBM79jmWfRLn+CQnOU4cy3Iim5KthTBJESRFitgIEiTWAQYYzL509/Re1bVX3dryvn/1nanu6enpnume6eX7gZ6qrrr3v/99721M14P3ez8LMA+f9YyuVNPnc0hH3C0DPQB6cLjFIx7p7vBBKzi4sH8qVzW6jk6W5DK6II6MWzKD4PYmRO5AaWUC16rRcJjr4wXVYtYWc7jY5fDsaFkuQd8P3xsy+WL2evVRFVAFVAFVQBXYjgoowNqOV13PWRVQBVQBVUAVUAW2rAJ+r0PiYY8pbyM4IsBy4o8wyuFYlna9UQA0YWRWCnDl2z9KyQ9+mkMJIXANXvPAXRUKIJgccGugB/+XOuSQIOoS2b2vii+CrDrcV/yHEIb5WTbksY/nZx3jBh+HdngkHIsaeEQ45QfIYpg9Swk/+1hSXn8vK5cuw5XG3CroQmcbg+sfPByWXQMB6Yi5EGzvhWOKcK+1DV1onfEmyi19BnDRgRVF98azY8jPuoTsMLizwgF+oUskygzLCOcambJkFGWNhVJN3pOCHD+Xly9/qtsAxA0uoS5PFVAFVAFVQBVYNwUUYK2btDqxKqAKqAKqgCqgCqgCd0YBlqax3C9bIGkRk3sVDrlMTlX7ithJj4CK+VV0XbHs7dJ0SV55OwfXEOAVXiPIaQW61ySGOTrgwuoHpPH6kA0Fl5eDW/BffNHJtGcoKG+fyprv7WPFw15088MOm2BEAJf4RRdZu2MtFHTK3XvCcuHyHOAVhMGgPgxij6M8k6WAi/PFqEk87DLZY8zLCgIgEiKyS+G+IYfsRXkgywXrAI1z6PSYhusqU8A1ACSbSFUkX6iZ7oWXZ8oynarJf/WlPundIIH4m+BS6hJVAVVAFVAFtpgCCrC22AXV01EFVAFVQBVQBVQBVYDghBCGQGUuW5Vz4xXxwUVF+FSusksewROBCjveteBTBe4qds374EIRZWwocQNwIdziXIRg4CxwCDlksNuHEjkPwszxAmZh6aDtRKLyd+3xm/yr0ekyMoUa0tvlk08+GkcpHifaPKMdXtHJxnM5M1JCpzqe99UxgxD3F9/IyuP3Ra68SPjF7oKs2CSgYjknhwflnZwXV8B8T21ZrsjhhD4lgK1crgIHmGXC88s4JkcZDrepmYp850dp+dJTHYBgG9/NZhauf6gCqoAqoAqoAmuogAKsNRRTp1IFVAFVQBVQBVQBVWCjKMBQ3pffysr3fpKC44fB4U1JxrwGJoWQT1Ws1A1o6oajh1CFDqEKMsprdB6RWAGuIPrJDIIWPg8Btgz3IlQcv0EWKwiGRyB8DKDMAuBhTR2DkGNhhxzcFcR+MQO/2kHQRtFmtesgiGJnx5EJpK8vGiwlfO29nLz+flHuPxgEzHOg9A+llQB8M3M1AwsZnB9AVlYVIItgkaWJvCbc1x587cieAJxvgIcfOAC9Wu+4AR6pPwo1zXtjyBLbO6QAy9ZNH1UBVUAVUAW2jwIKsLbPtdYzVQVUAVVAFVAFVIFtpMDlGUue+3EanTVrBpywbC1XKMnzrzvkYw/FDRQhlKGzKAigxdI2r6eJUr+gvHMSLix0ZGO3POSLCz1CsZBX7j0UQcc8ZDXBLTTQ7ZVB5GFx0HFUBSAj9PKhjJDAZSsNQikvfmu2aFFbYpQBpk7BnVUCFOxOeIwrrQxtixaoIAYdb5EQYN98OWHSh46QAFl5dCZkGSLnp1uObrZHjkTk+KmiTM9Z6HIIdxu0dAFihf0O8UB3o/MSa9CXVAFVQBVQBVSBra6AAqytfoX1/FQBVUAVUAVUAVVgWyrw3rmiFJChxJwqlsA1YPchLDmPfKXH4Q7yohyQLi0byRA6JcLovtcdlAePhOUnx0RyCBF3ImS8Ex35PvtkEvlWfgSSe42bKIo8LHvQZeWCw2irDpYAFgAAu9Ft8DyywdoHz7oH0CqbtyQdoBGtKcwgIxwk2KPJig41wimr6hTfPN1j6SC/6MJqB37sGPnUY3GThzWKDpCEWz7ArQ7onoh5pAPH0qEKqAKqgCqgCmxHBRRgbcerruesCqgCqoAqoAqoAhtGAUIOuqP4SGcTS81Yinerw+til8BWN0AX5kNh2vyULZDFb1zulpvKPlYs4hAnuup99GgMXfP8MopueE4grp0DXtnRH5Td/X4Jwzm03cYcwtVZ8vc5QLw//ItpySBw3VYzBGfVwWEfgCBC8uGYouapTN10HDx9qWAC8P2wb/V1ueWBA2Hph3OtfSy+1CwZHO71yd/4dLf88LU5SaWrEkSAfE/SJ/ftC0kyuv30b9dLn6sCqoAqoApsXwUUYG3fa69nrgqoAqqAKqAKqAJ3WAFWpKWzgCG04WAgVsq4dOKRW4cU9x0IybdedKOrHeoAwcMIRppwEvUDpIQCrXI1hrgze4kuHzqqcnWnKSdswGkURMnaHnTJ474u1BD2Ibh9O8IrXhc616jDjj6ffPHpTnn57YzMId/Kg3JJdh9MwhUVC6HDIOBfBQALNZvy1smSjE2VuLuEUaKZL7lRDgg4BTB4o8Hywh19DvnCz3dKCiH8dVw3BufHAMsIOXWoAqqAKqAKqALbUQEFWNvxqus5qwKqgCqgCqgCqsCGUKCIDCQbXtkLYvh32UJ3P4ClWxmEYM98pAM5WCnJIQfLgfl29Hnlow9EZQgQJYn3WUYI45cJcS+j8yCH09k03QkJtzhaGU2AMNvQeWUEwB9+5lN5HSYTbKDbLZ+HEysLF1YD5ZU+GKrYYTGIbUKAgMFmQy6N1+DCgnsNwMqDij82bKyhG2GpXJcpOKrCwVZ2mD3/Uo8sJeRXJ2CiDlVAFVAFVAFVQBXA/4xTEVQBVUAVUAVUAVVAFVAF7owCLBtcarAj4FqMx+6LSm+HW2YzNfGCmUQCbpPPxM6D9lhs6CG4qtaugjVUxRlAw7ym7ToIp3o6vIB7NVMi6HY1kQvmMYHtBWRdsQyQupUR4l6rwdHmsgzQMglYcNfRwcZRRtA9u0HqUAVUAVVAFVAFVIHVK6AAa/Wa6R6qgCqgCqgCqoAqoAqsiQJukA3CosWD5X5rMdz4TW8/ugrminUcp2FK3AhjlhseHJvd8grzHfL4fQjuq8VZTcvNsdXeY9keAVYYpZd9eKxU6yaAnaWXQ3BmsXSQnQaBsuTSRFk65lrllhZcV6w95HZ8RqdWYg3KQ7eavno+qoAqoAqoAqrAShRQgLUSlXQbVUAVUAVUAVVAFVAF1kgBdrRjnpIJbEfOVKUKuNHmxCIw8qG0b60GwVOrY+DKc7W4hnhEf01svwZgUBJBBlUkxFev1YaQiqMX7rZMri79nQBdVlNKVkN82DwU8MjBIZ9xwJkN9Q9VQBVQBVQBVUAVWJUC1/7tu6rddWNVQBVQBVQBVUAVUAVUgZUowO502UJd6gz5xiDAisLpREdOGaCDrxMc+ZF7pGPzKsBg98EuD8oJg9KFMkOG9Hs9TTm8Nwy45TW5Y5v37HTlqoAqoAqoAqrAnVNAAdad016PrAqoAqqAKqAKqALbQAGGtDPTKpWrSRNOK9upQ6CVKzTgdELwN5xYOja3AmwkmUFYfhUOO+d8ySEBVjTsvpKBxVD29Ro8PksWt3NW2Xppq/OqAqqAKqAKbAwFFGBtjOugq1AFVAFVQBVQBVSBLahABYHdM3NVScGFM5dlsLfLlOYxS4kgi7lUzeb2zpfaKpe9iA6DpmSwUjNdJB2OJnLDAC/RxtHjchlnXQjXfT3G869l5dXjOQOwehDa//QjSdk1cONOh+uxFp1TFVAFVAFVQBVYLwUUYK2XsjqvKqAKqAKqgCqgCmx4BdgFkF3hmEHlBGxYy0GH1chERcZnq1Iu10wWUq2G+CtjtnKbXCqH+YbHVQfWWmp/J+YqlBoyC1g5kbIMSOIaggh33zPsko74+v3K/cIbOfnWS7MAoYyJpwusClg6I3/7F7olEb/abdK8qX+oAqqAKqAKqAKbWIH1+9t0E4uiS1cFVAFVQBVQBVSBra8AwdXlmaqculCSsamylMoV2dVbl94+5lOt7vxZJsgSLtNtDhxhMlWRd88U5dyIJcRTnUmX1CoNSWfq0lVxS5WlXh6fKS9rQazVHU+3XjsF2ASSINPOJrvZmcsIa5/JVq/AK85TwjWfTdekCyDJi2ystR4wd8mP3sxcgVecn2WqM5mKHDtbkI8eja/1IXU+VUAVUAVUAVXgjimgAOuOSa8HVgVUAVVAFVAFVIE7qcB02pKfHUfp1Ts5OXfZklzOkkbTIT96b0T+my/3mlI/O6/qeutMZWryzqkiQBhBVVMiQQ/gVE0ujJfl8nQFJYIiiEOSVMYjPUkfspGQU1RzGKhl4b2Ad+2hRvtaCdYujVtShDuIJWV+//oer/3Ym+H5ZArlnZkqMqoc4hRckHrLxbTatRNeenChLYTxLx701pUqTQCsxe/c+veEZhbKUBePBtx/pdLiV/V7VUAVUAVUAVVgcyugAGtzXz9dvSqgCqgCqoAqoArchAJ03IxPW3LsdFE+uFiS9By6A4IDEEL94NW0zABs/INf7wf08c+X/F17kFS2Js+/MQf3VkUyeD4yhdItwJAZ5F3VABVcLqcEfA6JR11SQD4SOwx2dXglHsZjwgvY5QR8QOdBz/qUD2bgBvrad2dlKl0xi/ejnO3TjyXlnn3Ba09mm71C3f/8L2flJ+9mcH2dcmDYJwd2eSQWaEjFuONWB/pYCcrrGA25JVNohfW7ALQIQLO49gx3572w1i4sJ5bZnXCjGUB14RV0uOTgzsDC1/Q7VUAVUAVUAVVgkyugAGuTX0BdviqgCqgCqoAqoAqsXgEniAMhxgRcUpm8Da9a87As6/TFvPzlq3Py2SfiMtjjv+YAzLc6P1aRFDKPLLhg6LaamashA6kmxbK9ecM8typ1cXZ5EezdKu/qQDlZONiqUSzjPXata8DCQwcPQ75v5PqyZ7/R4395ae4KvOK2PNa3X0nLrn7AM4CW7TrqIJX/5s8n5fuvZqRRh+sK49JkWabmAvLoYQTro7TUdxNuqXDAJQnASoa2c2RxX1WqdQk3AZiKdeSgNWSgx4trvMr6VDPb0n/4AM1+7pG4ZH+IssFU2Tj7XKBaD98dlr5Otzz7/KxU6w45eiAge4YUaC2tor6qCqgCqoAqsFkU2L6/vWyWK6TrVAVUAVVAFVAFVIE1V4COmb4uH/KKBBCDzqu2gffAp2QUzqq5fAOlhIAQ88DJ3oqMooJ8Ixi5hBlKRcAplgtWFxlhuF0eQKuMr2TMLTt6/SYPCcYcdKxrwP1DV04LeDCDiWAsEb31X88IacYmr60hq6BD3unRijxw4NaPYWux2R4vTFjy6rtZac7DK66/CtD0/rmS7AfkaV2N1Z+V3+eU4V6fcdrNZtBdEjNFQi6ASaeMo8SUjQLScGc9eDACd97q51+8B+8Vdj3s6/TKlz7ZKWculgFMa3JgZxDNA6ry5f/xDEoXLTQncMizAY/8tae65IufSC6eRr9XBVQBVUAVUAU2jQKr80dvmtPShaoCqoAqoAqoAqqAKrC8Ajv7vHL3gSgcTwu3I1zyIXA7EQbkgTOKYe+LBwFEJOwCfHKZ7oVsAFdH2SBh1uLB0sQaIAKDvHs6kYOF4xFc4V/sjz/aRqFUl1nkahXh1iHwuNnB+Z2LT2x+Mvd1Xr/ZY222/S7AOVe14IrjhcagznTA0aHGrLCQ/+bpkg+ZZjv7/XJ4tx9ONx/uI5SPlqqmpJTHYanphfFrweJqNWSZ41wO60V5Ygn3CviUKQ398H1RANmm/Nuvj+M95LLhFmK5bA5r+OYLMwBbIK06VAFVQBVQBVSBTarA9v3fb5v0gumyVQFVQBVQBVSB1SpQQQTSS29nZGq2KqGQUx69Ly49sUXUZrWTboHtCXm+8skOdAosyLsIYq/hgz6hkh8woBeulnsPhMWN3CrPEt3jGPo90OWWiwkPSsVqEgb0SGNbsiECq/bBOUOo3mIWViLKXCwXoIlIDm6tk+eLcOo4pSPhkHDADQBG504LqpQqDklE3GbO9vlW8pxw5sCugBw7mV+weQTHOLTzJurjFsyyub/pTDrE53WhvK9m3FYGE+IPJxL29w66EOh+6+fHOQismH1m5seUDFznN5MzuF+CVemMu014/M0cLVdoALjVZBROq4sTZckAZoXgEtyLLC/eUwWULBLKtUO6Qq4mZ0arMtTrvZlD6j6qgCqgCqgCqsAdV0AB1h2/BLoAVUAVUAVUAVVg/RSo1+vyH789i2DxVpA3j3TuUkV++dM90texBp/U12/pt2XmSNgp/+SrQ/KN56flxTdHA8srAABAAElEQVQKUgDtG+wJydOPRqW/yyNBwCaGrS8eZZQPziKsvTPuhPMqYHKNImMeeet4XtIAWu3D64PTyuWR0ckqAJYls96qjE9Z8iN0P2zAteUA9Qr4CcR88uBdEZSgtY5HAEKHzeLyxfa5l3v+zJNJrK0pJ1FaxpLCrqRHPv9kB4Dc9v71b99gSA7sCMrbJ3MGKOEKGNBz9IDf5EYtp+lK32NYO3PGmpOtnzvCK5aXRgCQ2YmSbq9sHqHvcPERhq5mzKJRwM/ez8nruNcuoxGBD/dLCPcxnWM8BgxYsPihHYFNzuYnb4KkRjQGazVS67aqgCqgCqgCG0yB7f0bzAa7GLocVUAVUAVUAVVgrRV444PiAnjF+avI/nnpjZR88anOtT7cppuPwdss//vrP98jn32yIhMT09Lb2w8IADcWzoYZRnmU9QWRb2S7WRj6/gFcLmculUzHOifsNlFs9/jdQUGEt7z2Xh7gCfQA/9KJY7oMOmqAVsjUylqAZIJuhRYcOXDnACp4XQj4thAeDu7VD4iViBBqtCDWYjfXagQmGPmln+s0IfEsI/OjvE2HSBCg57d+sVf++HsuOTdq4To34LYLysePBqVQANRaozGIwPY5uJ6m01VcAzjsog4c2yluACw68Kbweok5anju9eI9wFI/Hun6u97Iw3n11omivIOf60vjAJO4riWWEVpwc3U4JJW3ZCccVomoR6aRw9U++jr98sAhJVjtmuhzVUAVUAVUgc2lgAKszXW9dLWqgCqgCqgCqsCqFEhllghlwgwsMdLRUoCQKgzYRFdMs+qW/m6vpAEeGKrODoH8Ylg2wVKpgs6FCOR+/0IJHQcBoQAP2HUwADBBIGUBSLD8MI2sI8IFGODAsZoym6pJuVSE48ZpgtrzxQZKBp3iwm9iDYAmNCBE+SJymODUYf4VHTwca1HO5mEZpMFxrfPVP0U6EKj/Nz7ZZbKp3HAwscyzVCoBUq6tOvt3BiQISDo1izwq/EM4FUGpH3PV+E++gFI/PApisWohXn+HxJC9xmvGUeO9h7yrVti/Qy5MlNBcoGzWzfuLmWtNwNYKrFdWDYvH8wACsf76JzrluVdmZRpdMqXpkmHkvf3Trw7j/Nb4BNdWLp1NFVAFVAFVQBVYVgEFWMvKo2+qAqqAKqAKqAKbW4HO+NJujtgadLrb3MosXD0Bhl3KVQBAygBgmS5vJsAd4AGEqQKXC8v6WCJYQDe5FlwAhIBDa3S6BggFUAR3Dbu+xZFd5cC2WcyVA6TIw17FfKtKDS4bBLdXMW/BiS51BiighA2p8C6H0zhxCCY4uCaCMR1rrwBBErsGImbsyiAkCvraXrjyzs0/8eI4DHOng68EQEowSSCaKwJowolXZwtMDF7rMuAoO1A64dYLB92S4X2GUHlWfHK/samazMwh4B+Q1IV7KIC1EqjWGrjvcBzCKQ/uox6UihpH4ANRsQBeWbYYQ6miDlVAFVAFVAFVYLMroABrs19BXb8qoAqoAqqAKrCMAg8ejsnx02WZTF/tPhbwueUjDySX2Wv7vsXcoBQyhuh6oeuKGVToF4iAbDqsGJwN3wwqs5hjVAFIsPAaO8vxvQAcVSwBK6MUsFpFWRcARRHfWDDBsEzMQt5VHvtEzXYAXGQl4FMtd41IH5xf+4cC4gIsC6DMjYDChmrb94qs35kT8pQAfiq4diwP9cLuViysLcDi6nmdu5NuyRYIrxom2H0OzkgcFiWGltCN58bxfbh3pgGoCvmKDA+EzT7MP4M/zEAsU06K+9OD+4PcKxH1ShP3XhX3KhsA0FV2ZG9Advb5AVCvlryun4I6syqgCqgCqoAqcHsVUIB1e/XWo6kCqoAqoAqoArddgV//fI+8ciwnE7MVk+n06F0xiYTX/oP6bT+xdTigBfjknW/SR7DBwVIvwqygvwkIVTeZRn50sfN4UP4FQFUF0aKjxwXQYPbHb1fFMkoFARZs6EDgQK+NCwCi4GjAEeOU7k6ULiL3yOd1y/5hvzxyT9iUm0UR/u2Fw0bH+isQgDOKXxzF4vppTkAWRwlqre4EyIJbD/eIEzDzwihKUeH2uzhWkQuXLXP/0HEVjeRlR49PHr4bICvuMaWobCpAyLUX9wqdVTTv0UXmwNz7hwPyyJGoKRVcf9X0CKqAKqAKqAKqwJ1RQAHWndFdj6oKqAKqgCqgCtxWBR67N4Lj8UvHcgqwRNCHsHM6sNpHHqVchFF0VjErC6wKHQgJFlAOhuyiAkrCitgmX2yKB/sjJ19qAFvYHBlHhGAmnshMiWIvlAo65eBwSB6/PyYDPQjgRvkgIYcP4Op6Id4sPWP3Q34RWuBf4wzzzedlta9Xn29MBViKyuvfbDrkLALkz6Aj6NhMWcZnrt5vDPOfTdelWilJB8LYGTrPLLQ8GgOw6UBfp9vce8xZi4U9cFz5JAY4pkMVUAVUAVVAFdjqCijA2upXWM9PFVAFVAFVQBVQBVasADvEseSL5WUEVgXkWxFGVeGaquMFlngRHtG1YzlrCHz3mdD1TBblhDW+B6cVcq/KcMjQz9Ok7Yr0Ct9wXjfdNcgj2tPvBoxAaSK+YiFErHPjZcb4TFXGpi3TyRAxR3Bv+WSoy4t8LZS/wU1nh34vM4W+tQIFWEJawvVjWafPy2t2gwuzgjkXb1LFfZTOVuUSwthnAaGm567Cq/ZtCUPHUfp7qBYQN4BnA4tjEwA69JjP1tPhle7EvF2wfUd9rgqoAqqAKqAKbFEFFGBt0Qurp6UKqAKqgCqgCqgCq1eAIIjB6QXkEplQdv6mBLfMDDoOnhuzjNPKgjuLuVY+v0OGOz2mNKwzidKwHMACMo5ojQLuQtYRu8618q/Av8wIwU3Th308cE1VAMayAFAlOKqCy4S1T6YsGZmoyGV0P+SxOSx0o2ugxeGu/oBxZHlI3jbo+AnKV//oOzMm7H4IbqHf+Gyv7BzceOCF3R8vTdQka6GtJIYf5Z2DgITXc8TdrNwO3E8jUxV0psT9hGsP7rnkQDa7AZv0bHXEPAZ2BrAm3qMsMWWjAB2qgCqgCqgCqsB2UkD/5ttOV1vPVRVQBVQBVUAVUAVuqEAWXQNfO54zneKcsFQV4ISZzVsyk0KZoOkqWDVuGLqwKignjCBAmyHaATCZCtw1FqAEbVcMfPd5m8bBVUeKuwstCgMIg4+H3cgwErhwavLOyTxcVHRyeYVwi3MGEfJuh7ezpLFQQjdDrAl1Z1fWzi6G+QKOAacXnUIbcaSR7fTsD9Pyn747hU55DMMXuMhKcnakJP/r7+wEyNs4C6fOE7PIM4OzyR4sFR2ftWQIWVQ3O3LFuszOVeGYQkdJXNuOmMt0Ebw8WcG1q6MhwLxTb4kDoNrQ5F858YQB7UmUrLKbIMsQ/XBkmQaWS+ynL6kCqoAqoAqoAltVAQVYW/XK6nmpAqqAKqAKqAKqwKoVIHQ6fjqPckAEswMUsMPgOJ0ygEVFJLSzK2Ed0IGcg194ahw6XrhqvIBTxZJDUE0oXrhkYij1ogOLMKsC4MRSwi5ACJaCzQFIzWYw90xFJgHGDu0OIAvLA1DhhXOrYSAXwRTnd2BHhsDTCWQhB6s1I6drGseXt3UQ8/pG+WMyVZVzl8vy3EspnDtADRbGqPQmzn06VZGvfWdafvcrAxtluQYEGvfcohUxb4ylpMygWu1g+ek4yj55L5E91jDRGycK8uaJrIxNWbiXmq2OkwCWzE9bPPYhmH2412f2Y+nqrgGfKW1dvN2tfm/h3uRaUzhGEcAtnmiYHLhbnVf3VwVUAVVAFVAF1loBBVhrrajOpwqoAqqAKqAKqAKbVoEp5BGx0yBHDbCo1iJIUkadVwUwI5vH+3DOMOzKBSxTD7pNmZcJdEfNl8ddB/gSCcJtlStXhIHcJbitvG4nwt1bbqo8XiDQYBmYC9VqzWbFlC26nU2ExDskHHSg62Edj27juKEzy4uvBsAWwRg7HdJ9E0SgNx057ES3kQYD7y+NV+TiSBHB44BXLeZmujAy0b7uaMpllOpR37Uuz7t5HeCEWiLviq8RIBIjrnakAD8n4OCaAswro6NlGmRzZKIMZ14NgMgh6Tzmxf0VRCkqr2sVZam8mXyepuweCMjd+wOmPNUNeuZC+8oZ7B8Nr3YVy2/PLK10rioz6aqMwhVWwr1ZrObR6TBoXF/L763vqgKqgCqgCqgCt1cBBVi3V289miqgCqgCqoAqoApsEgWILejCMulSTXSOA2yowKHFbCunEyVxDjeAAgK/QWXiUZ/pGNcAEJiYrcoEoMVsXozLpgYwUYeFiiCrjPAsfg9PFWCXS6JBlIR5anBhVZBx5DS5S+EgXFoAPSxry+RrmNstO7rdcmqkDpAhEvM5pCvhhTvHK0lkI22kkUHZ4MmLZRmbLBudfKi+K5Swwjb+w6fdnS5kgG0cgMVsKXaAXDwiCPNnd0h7EPhMwzH3zumCAZX3HgijI+DSv06/fbooL/x0DvlpVUA8lHvC6RQAuCK8IriLAz7minDqwWnncrqlA4HsUUDLOpL/eV+wUyFLRZ1OwC+49cJBS3b3r03ZJc+DYfU5ZL2NIo/r7EhZ5nIEWFWAtRZ4ffBweMPBUfs66KMqoAqoAqrA9lRg6b9xt6cWetaqgCqgCqgCqoAqsM0V6Io5ZSbvNI4ZwiuGrbvgngr4CKYsycE9Q1cVXjKurGIJ3eEcgFcM1AZ0yqEUawKd4y7CgVRCdhXD4CuwTDGQuw4YUQbEooOKwwUIxrI1diNkkHcVz2vzdiW7q2ABbiZ2N6Tpqxdd54hS/HDs0PGVBNjaSCOP8rPzY2VTMgg+YtxE+wb98mauYLKljMEJJ9AZ98tnn0waQLdR1s+19XW5USpKxxWXDrCErLLuxFWNGbb/J3+Vku/+KIWSP7ioALx6ABK/8tlueehI+IqDi7Dq2Mmi/NlfzqBUsExTFYAV3Hq4CUpejwx0OHH/EFY2xA+nXhLgqivhk4FON1x6AKUQj5ln1JANBThYxugD9GJJ46067givMrm6KWWdRXOCVxCyPw6IRZDF7ogu14zEIh75YqYqz3y0wxxf/1AFVAFVQBVQBTaCAlf/Vt4Iq9E1qAKqgCqgCqgCqoAqcAcVCAAOPXgoLG+fLgFSVAEzHLJ3MCgnzxeQQ4XAdEAElg4ScqAWznQcREUcAFVTTo9U5L1TeUnl6+i4VzMgwsI+fJ87EFLZ8MrsTuCFgPipdA1d+ZwAIghxB8AgOPPhsYKyM2ZuuU2GUiusnesJwS1UACxi2VkIZYQGDHHCFQ46u5h7xOEFoLvZMHDmenEuD9ZLR5FxGmF9XH8U7qIqHFb7dwYNcDl1EWWTEGL3gEd+5VM9EoHDiPttpOEBlewHxOrpCULThSWF5IrffnlWnv3+NJxlNaN5uYJST5zjn/7ltESjLTcdM8ouoWPkHzw7JWculQzsNAALHMoP8xTi/wGIEPiP+8xdwzX3uuW+/WGE+HvYvBKVqeiCCYfdmUssNcQNgsFrHoUTLIkgd2rqv/lMeTMfnVfMYWNZI8sbp5lXNl6WGTgHKwCstUZVJvH8n/1BWY6fKsg//K3hVd9j5kD6hyqgCqgCqoAqsMYKKMBaY0F1OlVAFVAFVAFVQBXY3Ar0d/ukB26nHNxPdNkUAKM+OF+EG0tQ9kd3Tuv8nMis8qKkzwkCdGHMkrffz8o4soSYlcUYLYIrVIG1ABFYBMxbCwbnIaJwY9JBdLrrQydCOqvo/OEgZCJUISSyB11ARXTHMx0Qq8jEAvRKwIkVAhBpL3XLAqKl4bKhGwibSBQOMQIjF774HgEGhwPJVFEcj3lcHHy5xKB6PM4AbLx/roCSsgbgiUsevitqcpFapY1wj3Gj+UGnEB1GJk8eEISAygc4xrXu3+mXLzyNTo1+tzk+3WUEdPzaiKMF9OYv8vwC6U76yTt5lIG2Mr2MfLjAVUCgkekqQE/JhPDTpffqsYKMIKSdrikObgvuBBddUyL+pikdDIdxM+HmuP9gSO7dHzTXqYTyVALJEjS7a3cQJX11lB0CVCIDq7vDY/S6WdjYWknrT/t24v3FgHlmfOULcAqitBXfXhlN1LE++/Ic7hCX/KOvDlx5fb2e8GeG9xYz5Nrv5fU6ns6rCqgCqoAqsPkUUIC1+a6ZrlgVUAVUAVVAFVAF1lkBgh5mFHHUG06JhJ0I23YByKD3HzKK+EGbH7IDBtQ45N0zJZlAOVYRZYN1fhDHF8iAgV0ESGxj5wQdIDyg04Yggo+EYuH5ksD+To8pE7TdUQQ8hD3ZQot8sfshXTk+OHSYXeTHe4RILEMsw6HDbKXpVF1OwfnDErFQEMcFJOHaOxG01AMIkkeJYxiQxB50UU0DunUhxykDsDUyia90Ud5FxtNfvZaRPDKtCMkiEYf88LWcPAqIlUw6pa/DJ4nIVfcXXT0uAD2uNwiHGEEbquXgOvIIu+kRcBWxDWEOAUUYutkg0F7LRn6cnavjeuHak2vhHHht+UhHFB8J6rIFp8nHGputAAxx+9amZnP+gRf6Ovzy2P1xE86+ezCA6+IGOAJEghg9HQCM2KbZdEsWDjsPstH43D8P+8z9Bm1vdfDehhXMXAsnjkt3F09nvnfBwumx7m+9nJbPPhGT++FMXI/BMsaX38xLFff3jj6f7EDnxY3YnGA9zl3nVAVUAVVAFVidAgqwVqeXbq0KqAKqgCqgCqgC20yBoM8tu9AVjh0IXa6SCd4moPLAbtSf9BmIZKE2EFzLDAawk1fYsIYAiA4neKSkDDcXg9hdgDvM0SLg2TuMD+344N6d9AhDw+1RtpqSA7xiWSGzterkWIQinB8f9pmXxFHAewxPvzBuyWzGkokZC/ADIAnHTSZcAHAegJEqoInbQCQGg7Pkr4T9XHg0cxRrOJYlo9M1GZ3Jyg9+lsX5gnJgAMnJbFaQkWSZrKShXj9Am0c+/nAcj1d/lYwDaM3MYQdkO3lhTSOs60TIPEvfOOga24yD15GuKLrYvE6UbzZaQJHXkRAuCfjXBeccL3oanQercGnRfcf3sLl5HU0rJRpwy+ef6JCH7w4boGfr4SbRaxsmfytEVx2AKaBfHZCSZZlBhPzfqgOLALQKYFbAvCyTTKD0sRv3cFecJaBti2h72sQN96ffT8kgrntXWyYYN6E2hHici27B1Qze3//P1y/Ld17Jms6ePtwfA11eefqRmHzkwYS5R3neOlQBVUAVUAVUAVuBq7912K/ooyqgCqgCqoAqoAqoAqrAFQUIDY7sCZlP6zG4mS5droBAiewbDKH0zw1wVAGQoJsKHeuQ8M7t6cLix3lmZrGzXFcHu8o1ZSbjNDCEqe6+gMgB5Gs99aGEAULMo2offjiq6vV5pxJoSN7RMCCDMIRzcdCBRchBcDKXt4ybp0ZyhDK+DKCaG3N4PYQWDKZHNhecPQQtdEyxPI3d91iCSAjBsre3T1tyZgyd6BDmjbfF8ANsz6PRdZYtoLviHIAajlVDjdxnnuiEsaiB84dLC9Bl9wCcXAggJ+BjllOwze1lFrzJ/hibqsKJlpJTF0so93TIgd0hOYWyyhLqQV1w1e0fDsnHHozAUeYEPGTAfwsKEhDCr2WcTdSQZZqffiwhTz8eA+xZGZQxWWJtQPNWpeO1pzOOIwhYxLyrbjYGwHqajg45PzqJxgRLHAXLreC+HpsuozS0lWnGregcY+mhXY7qQVB9wORztXLQSiivrOG+onPMLlGtAFrNIvMtC2D69R/OyktvzJn7jIqwicFFfP3gp07ZA9deGMdSgLXE9dCXVAFVQBXYxgoowNrGF19PXRVQBVQBVUAVUAVWpsBgjweOqZAM9/uMKyqIPKcQHDEs3yuUmzI+45F0Bl0H4ZgC1zDwyonfslg6l0Rnw8MAYAnArxrAlQWIQEDUiTK8o4dCKCvzmtK7MOZbPHiMoMmXako/3s4CGOQAiBgSz0FAYNxUKG0kFyGMoBGmDsBFXw/LEfmc77EEkSVqBBnMPuJgvlcTdXGEUW+fYOkgXFxFwgTztsnCaj3j/oRY/K6BLCiRDy6gbHLmsnFW+QGwDu8KymceT6K0cGv8esnufL+PMPbUXLmVFQU71UCnVz76UFRqgDf7doblyQciCG0HOAQcpNYEgkl08HNDZwItuqeiIY986J4Iwus7zPWx9bydj7xudDzZg/cDuxnyceigFyAuKH1xp/xPvz9h3H32dnzsgIMugXMijrNzzwgz2+EVeeoMSgF535iy13xNwoBvvBd4j7EhwRxe+/6rGclm6zKTs2Rkoma6cpLy8c7nHBZ0nYKLcAYlm3hZhyqgCqgCqoAqsECBrfEbxoJT0m9UAVVAFVAFVAFVQBVYWwWYQ9SL8qYgysDm/CjRAyUiWErB+eRD9Vh30itzgEsplN258GGd5X1x5DwR7HThvTg+zLMErRfPI3Dr0EVFx1VXwmOgx6IqsgWLJ2SwnSjMK2LpGh1TdL54sQ4Gi3vhtArUXABadZPVRahmQuexNh8m70MwPYEA4UI05IRLqlUWx655PpCpn76TkddPFpCDBZCwlAsH+/K4PGcem2WMcwBhhBPM1EqiVPD46bzZ5nNPJhesfzN+w6ywP/3+DKBk2YA/nkMDtrJxlGcePRiRhx8OmVB66nFwV0BGJ8ro3FeTBHTw4FpYuAfokquhCnO43y+PAmD5AT3tQRjI9+3rar++Xo91ECzbtdd+DEIjNiGIRUSeeCAg//Z/6JJ/+K/TMoNcNILYbpQMHkagfB/u/TjuWztcvQowZzuvOB+7ZeZwP9BhFYImdGdZc8yDa8FTuqu+9cIcfj4q0KQJKEh4BcAHHXgcO1vMlN9iUUH8/GzUkP92/fS5KqAKqAKqwO1V4OrfpLf3uHo0VUAVUAVUAVVAFVAFNpUChFheuFTYfZAf/DkYgs0yOeZcMfeIZWVulFElEJrOIGrmQjEvyo+gdmyCEHaUmKGmLBQA5MCH/AxcKew8t9JBmEVgFOpvBaXT+eMuOmR3v0POjgIqwBnTQPB3AN3uCAD2DPllV3/AQCs6rwhMrBq7BToBY5rI4GrBh3OXLRMQj90XuK64Li9+W6TLqzPeCoonCKvASQY8AxAHVxnLGHFeDDmnG4nnZHdSXOl5bbTtpmdqKJWsmutsSkIBWqhtFRDr4njJ5JbVQugoOB9kz66VuJxy5iLC9VE6CXEAaJhf5TAdBBn+z+vN7ydScBihjK6ALpe89vt3+E2Q+npq4MZ96XC0Oii2H4f3Iu8pe/QmHPIn//M++cbzabmAUlk3ygK7El55AN0SCWDt3K62XcyuhJqEV3yd9xAHSyjTcFsxZ+3CaAmuqoqBXnRptdyArZwwbmvPB+OacSQ+dCS0YF3cRocqoAqoAqqAKqAAS+8BVUAVUAVUAVVAFVAFVqgAYQYBEh06zLliKeAM3CQvoGPfBLrPMfvJDwcKS/L8ID8EOezKxxwsulcYrm4hXMqFcjSCDgIw5lGxVHA1g9CB7h9+RYIo58J6fAhOZ4A7nTE8LsPW45Gr87KLIjsOsnyQ5xEEVKuz4x3yrLgmng+dVXTEINbKwBs6vu7eH5Qj+8Io+SoaIMG8qwLKGGdwHJYr0j3WwPlWqjw3dkVczZlszG1Z+kl3G4eJ38dT6kq3EsEgA9VDcCQx3J268rx3Qu8qIM7IZAXldbzGTpTfOVtOLbiyLl4uI2esJufHSgB9yFFDKWYUuWEnzgcQXB41eVTmgOvwB+8XhsLzPrEHX2vvSGm/HobL8EtPd0gq0zBZVfyeGV+EV9yHg+5BlxNAbp5W0d3FLz/udd5bHHRdpTLspiiSQpMBdrzkP5zCD5pbBehFgpbZjw0QGHh/CBlj//1vDJr5zST6hyqgCqgCqoAq0KaAAqw2MfSpKqAKqAKqgCqgCqgCyynAUikLZWGEUSyPKgFY/CnCqKenLQOA8JJUGGxdr0oPSwfhwtoLF1QHSrEKxauB1ywxYzYSS/ro7LmVQXcUOwzya/egSdFecjoGaffADcZSQOYXRYOtHKRcns4sp3HOmA6HABCEEGiyKDsH/fILn0jK3sEAYEVMzoyUALxqBsCUTgJ6Ye08vu2h4fxhlChu9pFAZ8G79wXl5TdrgJFwlwG7eKBREoH3Dx6JmOB7C9d+araK8k2UcJpyUIfsgV5xwMMJgkI4riKAncmYy7iTLsKdNoag/HdP5dBVEh0tIVMmh2B13FPsBviZxxmovn7KEcoRoNIpRX8UQdz1ShjpJOxK8uv6HxVicJ/lcU8TVIUBaOk4s8PaqQfLEAmv6KpK4vzceC0F5xkz2ejqY+ZbMuJFYDvytQCwHjkSk499KLZqmLt+iunMqoAqoAqoAhtNgev/rbTRVqrrUQVUAVVAFVAFVAFV4A4qkIWrhmHU9kjlqibIPIPw6lZ2zzzGAR8ow23CLmw9Sb/JD2IuUCiALoQoSyPIcqJ0q4icKpbhRUJoYbjGowBX11QKnQQJUeCe6Yh7DVwgsOhFuVs9gfIuuGPK6KZHAEFwQqDiBnDgc1QIwsXlkV19bhnu8cFN1oIdj98XlVE4jBhW7vG45eT5Ipw1NUAcJ84zII/eHUIm2OYHWD6Anl9/ppucB50ZC8i0YlmoS575WFJ29XgNgMoC/LFu1A+Il8a9kEWAP7WlM6vXibJRzMESPZbXzSEbbWoWXR7PFFDqCdgJ55EH70UCvBcE7ray6Sa5Wifeam8bXn83ylfXYrAhAMtkceeY6cLIfJvEPUc3XhTPKyijJdw0Dr2GBXefA44sBL3TBIYlELg+82TUAEFmwyURFq9DFVAFVAFVQBVYTgH9m2I5dW7hPdqoq1X8L9pNOrh+i7+tbdOh56/Xf7ve/7z3t/P9X4fTogESsV2vfw21X9v5/Hndr3f+zHnKwXmEjHQDJAh52E1wJlVuUSsyAXwxB4uPBB8sQfN4anDiIOeK7hdwnQbuMX7vRBbSdBoZQwAKHlfD5GSt1V+5dPUcP1tAxhK6DoIt0GnTjxDu4V4ABRzPHkFfQ05eKMocuuwxoDuAMjcX8r3onmE2044+H/bzATQgIN7bAndcfy+cOdKE0wo5W7v7wwaE0UW2D06zIAxguUIFEOvqcezjbfRH/t7Wfv15Ln/3C50of4ub0kCCKDrmGGBewDky1yqEbTJZAiq6kJxSx8/QIAAX+JQpn6P2dKy9+UFJ/ur1nCkzLRbw31nI4/Uw5JyZWU7T2VKacG1Zd043/vzzv4E3+98/P+DnUBcD3NEdExphKvys0KUnKJ9E+SngFX826O6j6bCETLafvJNHvlYAPx/umz7uWt1Xt3r+a7WOOzUPz59fN3v979S61+q4PHeO7Xr+/Nnf7r//beb7nx2IPbS8boOhAGudLjJ/ACYnJ9dp9vWflv8R28zrv1WF9Py39/Xf7D+/t3L/85eX7X7+/BC7Xf/7Z394387nX6lUlrz+DJ6eTsOxZMqvWj9lZbibYgjyZlQSPShEPIZf4Q+6TjrClhTzWZmcysnMjAMlUw6AD/wPArijanCjUO86oMiZSw7Z1euSgc5WLtat/gy/c7oqYzN1KZa4HkCWoEPSaZc0Kh64va4Ckhl0iZueQVA5/s7v7QSgK2A7rC/oYyc5uGj8VYkFizKL8kiCm/bBkq8IAJgL8zuggwPnWyqV5MwFhwmGTyB7a/E+7ftvxOf8u3+5n3/kmUsJ+pgsJ7jWGgCY49CQ5aCENFXAp0oRAC+H8H6UUdogJwXA9dJbRXTea0Cj+ZB8zMVSVN47xTLysEC0UrNTd1QW/rd/LX//4X2eybfC3U/AqUcHI/6KMYOlqkW4FE9eKMi/e7Yin/lwQA7suLMfS+zz38z/A/pWbiCe93YGGPb5l8v4nxLbcBDcEYLwv+PbcfD8ee6FAlr0bsLh9Xqlp6dnE6589Uu+s39TrH69m2YPEtChoaFNs97FCz179uymXv/i81nt93r+ev0388/vau/39u35y+v58+e37c8/P7yNjIxs2/PnL/Dj4+Pb9vz5C+zU1JQMDg62/1iY53Tf1JwVk4dE94jtZOrtacro9Iy89UFOnIAa/IDuwoeAg7sCcv+huAzCxcSgbA9IRSdKpOZQdpbKoIwQuUdnL5XlZyfy+L5mnEtH7wrIL328U7qXyR26ZmGLXmAOUflkSmoOS9xwSHGw6rHSRIlWR0x64cSyR6VZlp6mJYEQgshRAudxl9FhryDBYFB2Dfjk0K6gHNgZQBe66/9fXUIJBsMXUYr49mmElCOgnKHmTz+SkN1DAftQm+KxWCxKJpORvr6+ZddLt1kqW5PLyLMaBwCsA0TOIaQ8AxdW3QFwVwfNw4eJAEL6o2GnZEoos6xYAFooo5sHOMSBhFdhOPM6EwH56ENd+LmLLnvc9X4zn8+bD29r9SGoDy4sllnmClUJvzklHi8/GNZQWgvYi3uSUoB9ysUJp/znFxrym5+PyyN3R9b7NK87fzabxXWqSFdX13W32cpv8N4nxOvo6NjKp3ndc0un0wbgJZPJ626zld+YnZ2FYxfwPR7fyqd53XObnp4Wn88n0eid/e/wdReob1xRQAHWFSn0iSqgCqgCqoAqoAqoAksrQGhFx1QWkIKupiZeCKI7G2HTP/7qoHz9+ZS8/m4eHwAbsmvIJ3ftDkocgd/s/MbBIGuWTjE3yAIMm5mry8vH0LlwGjlVsO/MAXrN/KhiQr3//pf6lg2yJqQqoJtcE66dCJw+7ZlTzLMqoDPe4mD4OQAXD0oE24cX3+OwgCyuVgYWWg/6vBUZ6o/IYXSD60SQeQRZRssNE2QPGPdn30/JmdHiFUDz/sWy/PYXe+SuPaHldt+U7zHA/2pHPrpWWxlXJWSaufEBMFtsSmMKIenICcsVHfLWmaJMpODGg7HDBlgmJB9llr2dXtk77Jd9w0jU32KDkJe5Vgx7f+zesJy8VJIKNZiHVwRYrNrKg2jVpuryh9+ckmnkyX3uie0JULbY5dfTUQVUAVVgXRRQgLUusuqkqoAqoAqoAqqAKrCVFCihXGwSH67HpyoAU37xsZ4MbpJcCQHpMYf86me65Cuf7jTB7BfGywZksSMbARGxkd+EnjNM3YkcpKoJQmcHO5afNZE7RUcKc4NeO56Tn74bkkMAYHQy+QCZGKBujxyCwk9eLMnkbCt83Y9t7tkbMFlV3MYJB1AUHeFK6HJHMGYGFhHCa1WUbNEx5UU5IHOXCBfoGiqUGCSP8kWEskf9Xnny4Zg5dmvn5f90w0Z08mIRweTlK3CGe5Tgunn2hfSWBFg8P5ZHdsKZZgEYngasKyEPK51ryGy2bjru9SQBuXBtJuDQOnaiCI2519XBSxPwISB/ICBP3h83XQ2vvru1nhHafuJDCZPL9upxlBTCsUeAR5iHW87ktDXxPcHsT98pyM7+gOxDN0feozpUAVVAFVAFVIF2BRRgtauhz1UBVUAVUAVUAVVAFVikQAEw6F/9xzE5jg5yVbRQ83pccu/+kHzsoRjKBRlaXjclgnTmhJBztXcwIGPTZRNiTYAVRJmYvw1E7Rv2yavHWmVUAnjFPKk6Ps1zW5bi/QwQy4So9/uQk+TAPMikQhliHa6vC+MVlPlVzHMu0wKAev39hnziYbi9sA2P1dfpAxxA0DghFgALRwxOqjTK3OjM8iCknSWRhFY7UeI4OYsSQ3zPAHpfsrxieMV5g34Huh3WkekFEGcO1Ww5zQB4puE64vGoy1YbPCc6jBwAhoSb4zN0WOFCItsMXEpKQbj1cF9cQEnlbHbppjj74Lz66ENxeeiuredSW3y9A7gvf/WZPtk/nJUfvpmVkcsVOAUt8/ND6EogiB8r3C91o2Vvku0wAX7b4O3iOfV7VUAVUAVUge2ngAKs7XfN9YxVAVVAFVAFVAFVYBUK/NFz0/I+uvrZFqOKVZc3kF21A4BpZ5/fZPmwLJCQioOuqeFev5QQYk3oxPwrgiV7xMJuALCgvPBGxuQnsQSNgwDJgxI0N347o0urD3lVzFEiIOH+iGiTcZQccs72UcZ6xmdqsneoVQq4F4CMhY5VOFpy6PRWg/OqI+E2gCxXbMwHuTfgEHKaEsRIKGDmrFoVmZ1dHWyiu2agx2dKJFHUiGMQWDmNJsGgZ0vCK2pPeMWv1FwVLjyWbLacdjVcm0mAOxc04LXDJTDXFZua24eMj3cCyzcfQubZfnRu3C6jB1DqiaNR5Kr55fk3svLi6xkAWgLhllsthFJWNygWzIpmlHHfK8DaLneHnqcqoAqoAitTQAHWynTSrVQBVUAVUAVUAVVgGyrAsr53z12FV4QzhkTA4XT6oiVPPWIDH0Il+3kLcNDhdL1xaE9QnjwakR++loGrC0FA2NWHD+9DfV6Jh1GahnRruwKQpVYN5F054da6eoTWzOxiyCMXkXvF7eji6ox7DEQj7IrAhUXgFQq0ABq3JhjwoE0inV2EMBUci5laFTiICLhqAGXsorjScdcOP+ZzSy5fwRpZLtlEmaJTHr/3zgVyr3Ttt7JdEJrSOcfSTJZlupxuU1LILCxIKwlkiMUZjg8XEYpErwIsXCM617pjfpQbbp9fxV24NwhvXbjnPvNEAuDKKa+fyLVKWqGRDyCrr8MrOwCFOXg/t4+t6uZrP0d9rgqoAqqAKrC8Atvnb83lddB3VQFVQBVQBVQBVUAVuEYBOoxIiEwZHEBQAw4j88EaL7tdDQOACI3ag9SvmWSJF7j9L3+yGx3+fPICIBY72XV1eJBl5ZUQwBeBCD/Qc3B+fvgn5epB6Hc6X+U36JjGnCt0skPpoA/A6P1zeZnNNU3+VTSEnCvsVAFJYU6TH/M55x1iNHCxXToqt0wHwcnUfCZRvQa3liBIvmYC6FvHNIe67h9FuMzeOl2UBw+G5NhZdp1rsAGfPHF/FCWWW72bEwP5kd0EeMhrg0sqQVy7EM4/geBydhi8d19YXnuviDJNlH0CdJHJeL0uOYRw+wcOB03p3HXF3YJv0G2VRPBVIuKWr/6iXx45HUYJbB5ZbHUTaH//gSDy2lr01DN//+fhIvzj783I2ZES7lmHHNkdkC9/unsLqqOnpAqoAqqAKnAjBRRg3UghfV8VUAVUAVVAFVAFNqQCdBaRLxHGrNfg1HfvDyPPidAIcAaQgsdlt7mH74oagMVOfatdQjhoiJT83IdicJx4EepehWuKodZw7zSc0g9QxVwgzhsJXf117TA+vFtWU0YmS2IBmsRQdjXcC6gFp8/P3s0iH8uSmUwVZYcO6UJI+5F9AQTHu2Q2XUM4th9lg63gbJY5XrpsyXlkNOXhvmKpVjjYFBfoXANfLFsMz7u2ltN2BOsug+KEAMw+fM9Vx1VvB8oYScq28JictUwJ4WVoUIQGDNSn+62/1yOdMa8M9QZMWekXf75LvveTWUmjE6QTF3QIJZe//vkedLDEBd+mw/55uWdfSPjF+63VWbMlCO/9EAAgy2v/zdenAa9yV5R65ZgFaNiU3/yFniuv6RNVQBVQBVSB7aHA1d+Itsf56lmqAqqAKqAKqAKqwCZXoAIny4/fzso3XkwZKDAMAPS7v9IH0HPzmUssFaSbiN3kWLrX7j76m5/tlHTGknfPlJErVZMggNXTj8blU48nVg2ubOkJ3VhiSJAVC4dROliTLJxPdHfxdZaWsRzPB7DEMj++zg/9fP7AoSDgiFdmAapa2VoO5GnNoRtgWUanSlIC4KLbJ5VhmWBD7oc7yudrmGD5XejwxhJFulrOjBbl8pQloxOWKSPs63TLUHdD9uD8VwqfSuWaWWMVdYdcoz14fi6bUtgvbqFHlni++HpW3kIWWgHB+8w7s8A4owCKh3eGZBDXJ4rryPHI3WG5/0BAJtMI+8eNtXcILju45nRcVSCAjoN0JbLpAF2PbgAsjtHJipwbRQnvovHeuZIUEJJPt6IOVUAVUAVUge2jgAKs7XOt9UxVAVVAFVAFVIFNrwBB0/d/nJb/68/HUSrH7ndiSot++m5Bfvcrg/Lh+8Ircg61C0EglivUUJJXB9gB6Jm0JByCIwmQgSVPbpTn/Xe/NigzqZrMADLtAJzgB+61GAQ9PMZQt1dqyP9hCLopW5yfnG4vllcRmJAHtZxSLomjRI2Ai4NrHyGImiyb9c83HoSbqi5jCBTvS/tl35AHjha39KBMsYy5zo9W5L2zJXnprYzZX5CxFcT5Ht4pcvRIQ6JwmK1kxFEKRvDQgLOrZg4McAXI1onQeN8aabSSddzubSw0FnzjVEEKAHg8Xwbi49aUYqEpexDMPtDlQ3lnK/eK5aButwfQ8XavcnMdzwa27atOA1I1W+0t218G6KpJBj+rCrAWyKLfqAKqgCqw5RVQgLXlL7GeoCqgCqgCqoAqsHUUqKCt258/n5LiPLxiZz6OQrEm/+G5CZTPdQidRCwx6gQQOoKOZ4QLyw0Dr+BaIryaziLgPFhDWZzDhJwHURZmB6B3Jt3Cr/UareD0lvPEPkYmVzcliwRYLLPC53aEpDtMXhDdWMxfmpmryeR0BTAFQA8QhU4ofjUwVR37ZbIW9nWb4HDOm5qzZATbv3Isa0oL6TiDxwt5TjV557QYGDaIMreVDJYvTqZ86IKIAHccD4cFXPPIEWQ88futOqbSKJ0sVc35EizSMeSi7igzdYL9sZMkwaeOW1PgAGCgD20JCZfbRwRBY/1dnvaX9LkqoAqoAqrANlBg/X4L2wbi6SmqAqqAKqAKqAKqwO1VoAmnUCbfKrVD7vmCQZDz6jsZ405iCSBhyrETAXQKjJmSOzsrqwm6Q9cVIU8VxIeOq1S2LlMplCVhbq8P3ftQmsRyMEIjAqN2V9SCg67jN1wbHVh8JFxjp8EKQJvk8S8AXg6B6R9cYEA4XECAKMwNokMNEglNKwRILrij6PAKsMsbwsNZGphHl7wSSgiplw0AqRX3qaAM7jRKER+9+2qe1XKnyHD7h+8KyfisV/KFhoGFAwALJvR+uR03+Xu8v6IoWZ3LWcgMa50MXwv6PeiktzL4t8kluC3LZ6fHzz3eIf/lhVnc23WAVjQnAFT+xY933Jbj60FUAVVAFVAFNpYCCrA21vXQ1agCqoAqoAqoAqrAMgr4UZYWC3hkFi4iQ6jmtzUABu6XIlwxdDIRPnFMwGl04nwR7qIm9qkihwj7wZrEbn7sFMcOfAzjJoQoWyjV43YZQCyYO6pwLxFcEfIQTtzuQdDGUYbrjJ0Ey4BpHHRhvf5eTk5dqiDQvW7eL8Bx1hEPSBq6MFRdcPoBn1v64Rjb0e8zXwG/w8AwgrBUtmpg1vwhTDmifY5WxRxmxX8QDDJ0XjpXvMum3zAZcyHLyofcMZZ34jqBGnogwb0Hg+b+2/QnuIFO4OMPx+QguznCMcjOnw8diUh3YvsG4G+gS6NLUQVUAVXgtiugAOu2S64HVAVUAVVAFVAFVIGbVYBw6m9+vkt+799bUqijhAvsAEYj8aJcqycJ54uzCSfS1dlrgFpj0xa69MG9VK2b0kNCqgvjJdnVHxQXuv5lkTHVgP2IAdIsPawACgFHIMMI5XqYihlHd2LQUUUnE51mXFsBzqlxwDa6p96H86qGBbpAnRhX5YAQBFTDh8IyMVOWBtY+hK6DjyNAPB71QJMWgfPCkVXCPB64soLYPg9yZzTk2zhNdkJ86N7wnTjdTXXMcNAtHzkalxjKJUeRP0b72m4E5H/kodimOo/Nstj+hEs+/9HEZlmurlMVUAVUAVVgnRRo+xVvnY6g06oCqoAqoAqoAqqAKrCGCjz5YBS5VA75P/9kSsZTFQl50ZnvYASlcTXjmvIjM+fKgDOG7qpUtoYAdGRCAQbVUWbHUroxgJ542GuAF5KLAIGQMcW6O+xDBxafMjTddiZdmfM2PaGziZ3sisi2Sl1uyHtnCoBqgHAAUNMzFrohOiUAYMf8JQaws9yxM+6S/UNxONCc0pXwSALdDJmVRYDFwPUwdEOVG/Rzy+FdYWE3t0oVji1oxND6o4dccs+ewG06w9UfpkaHHK4lux8GUEpmh8gvnonXmSCSg50c17oElLlXdLbhtpG9wwHxQuMBBPGze6QOVUAVUAVUAVVAFVgfBfRv2fXRVWdVBVQBVUAVUAVUgXVU4OhdEfmDfwJohfD2y8iAslBmV4R76g2U1s3MVQy0IgAi4EjgK5VBuBMGwYNdNlcH5XDOAyC+5/eg6yBoVQTlh+EQAtvjXgkDEq3VYJbVFEoWuQDCJS/Ayo0GXVh9CKN/47083GLgJVhfDevmc7rFIkF0uMNrdKbFsOb794elF9szw4tlh1VmfeEfBtETZhHe+ZGF1duBLoF3BeHOQhklQB4nPLrfJx+9r6XRjdZ1J94nvDp9qSQZ05muCcNYDedUl92DfgA6XNj5cWHMktdO5HHuNZQ2+uQwgFwsjPNdgd72HDd6ZDZZHk64JO6t5HxcWOvewrr4RIcqoAqoAqqAKqAKrLkCCrDWXFKdUBVQBVQBVUAVUAVuhwLkBBFAmwP4sscQspheP5GTKeRdxQFxdgx4JYgsqHdOs4tf3YAOZkjVYM5hRlRXzGMC3RuASlW86MI/uwb8sqPPb9xX9ry3+jiD9bx9qoD8qlaOlRew7N79QenA8W80CuWGcVt1xD0GmtAV1o8ugRPTLSjngfsnABfa3YBXTzwQNUHuBYS++/AaRwCONBvEOQF6OjEP3VrOuMNArWo1bNZxZI8LIfbpGy3njr3P0PkMyj0J5Dj4yCyv6JwbnSdbOp66WJJvvcTA79Y2I5MVuQy32lOPxtcUYOUQWH/ibEEmUjVo65KdcGNRV0JFlnLqUAVUAVVAFVAFVIG1V+Dqb3xrP7fOqAqoAqqAKqAKqAKqwG1VoL/HI8/0JM0xmR3lcDSlUGLnvYa8f7ZoQqDprmJm1iAgEJ1QbtQKMhPK5QAgmSvKcK93TeEVF3P8DDKrAK+Yv2Whk2AT5YqvHS/Izz8au2F5mwdshi6rCNxgeDDjXriKuqI+05EtiKCuvUMBeehwGNvRkeUQH86HZXQsg1zcEXCgxyvkO4Vi1YC8MNxZ3XBt+bw1KbSmNy41OowYUM5/WJbJ7e7kYOdFrmXx4DrRg9GUhf7s/fwVeGVvNzJZlrGpioF1N1tKSCDGfQlN6dx79gczcnqkiIQ1mNfw4plRn3zk/pjsHmorX7UXoI+qgCqgCqgCqoAqsCYKKMBaExl1ElVAFVAFVAFVQBXYaAo4DW9xwCHjkKMHQjKIjKLJqaq4AIT6mFeE8i8CiQqAUoWd/gBrqnDPRNscXWtxTjmAF7vDod1JkPNmAZDGETBPkLbcCNBxhVK4i+MNgCRpdb1zu+G28iP7K2Q6CC6GVCxjY1nhUiOATo67B3wIr3cZABNASWHQ7zQdDe3tsyjTs9CFkYOlj/miJfmCU5IxwD8Ewd+OwS6MhHA2OGoP528/PmEdB9dZgvPMHtyf8V4s92PHxo6Y15RSspMlNVjJKOPeKACQ0aHHe4VuthPIDTs7VjLacY463puYKss7Z9yyb5ix/zpUAVVAFVAFVAFVYD0UUIC1HqrqnKqAKqAKqAKqgCqwoRRg0PlAl9d8LV4Yu/G1AA4yo0otGLJ4m1v53mOcOwBEpCltA7HxxiFFeLZcPhNh1N5hHzKzxGRocb/uDo8pW2ubblVPmRnFksSlBoGPDa+4NjqfWHY5ifyuuXwNQeUe7Ou+oXNsqblX+hpBX6HUWACOupNemcu1SkEJpzhYTtmVbP06S7AWgkstW0A6Fs6hQvcYNmMOGKFkvlQDzHIZhxbPkaH11xvs+phFmWAagfF0vQURGE8gykD9s2NlZJHBjYXJ55dhHG1z2JZOPh2qgCqgCqgCqoAqsD4KKMBaH111VlVAFVAFVAFVQBVQBYwCdPx0Jz0AIgsBVide8wC6MI78RoNlgbuRzcWv9R5NdGG0ByFSESWY03MWYE0T8AdgCS4nupHoxrrZkjx7/sWPFbi+ijgmHWCt7oGt0H2CIwbT0zk2hdypSrVunGA90JCuKA6u6UNHwvL9V+uSQ7g/PVZNvDjU48dX61deAjnCSgKy6wEsruHS5QrC4msoS3TguABiKCFkEDyPQScY3W0OBwL1AcfMwOuxCMpRcZ10qAKqgCqgCqgCqsD6KKAAa3101VlVAVVAFVAFVAFVQBW4osC9+4LIThIZQ8mgA8iqCw6mPUN+A0SYqVSA84tOHwav364udgRShFWtUssrSzVwhiWJdCGxkyHBG7clrHKC4PA15k7RlRQKrB2w4ZwMly8ir4yB+2V8MaTf7jDIEs8oIBZBVnsm1dWVCzQNoGTUjSD/vIFgvXBtdSfpnlq4TjqnmJG2+Nx5jT44XzJOqzJgV8uB5kRIvNe40gjV9gz65DKu42zGEs98UD5D+R+5Z74dYfuC9LkqoAqoAqqAKqAKrJkCCrDWTEqdSBVQBVQBVUAVuHUF0tm6nBotiFUWObwbXeoAOu7UKKB07PiZkim92jUYkN3otKbj+goQihD6eBAQT6dO+yBAeeBg0MAPu0Me++jxOYGNPUoVh8Tns7ns19bjcQrlgOzo10ApHUFU73wXP/tYBEWZfFMaADrchusPwLnEQehFl1QeLiefd2nXkV3it1IYRx0Ir1qj5Wrin3Re0fm0eCzndOpBeeWnH0tIKgMHFS5KKlO9Ev3OUlIOF5xSi+EVX0/nAa0YvIVBcGZZ+IIji64z6sTBwHwXdn77dEHSOAZ1efyeqOwb0p8PI5D+oQqoAqqAKqAKrJMC1/5GsE4H0mlVAVVAFVAFVAFVYHkFTl0oyf/9Z5Ny8lJRKvjg7na55HMfi8nf+8LA8juuw7sTM1X5+g9nZHLGEgtWFcdP5+TeAxH54lMd63C02z8lA7/p6CFoYo7SUjBjJasi5CuWm/LWyYK8dzqPoHAEee8MIMw7KJ2L4CMhUCLqNmHjYEIteIX924cNcljmtl5jZq4mqWz1yvSFck1GJhrS13ZpTUZWjJ39Wuuk64qD6yMUKlmCYHonMqnqpjshyyQ5CLtygD3Ul4PldpHgtQ4o82bbH/PMyLzCLKtyhf0GWwDN3swOa7e/X+6R15VOral06zxLuEYhdFEMQVe6yLimpYYF15Ubrqo6zpOOM8IpgrUKzt9kbCE3i9fmyN6g7Md1ruE8GQi/GFguNbe+pgqoAqqAKqAKqAK3poACrFvTT/dWBVQBVUAVUAXWRAHClP/w3LQcP52TAp7T+QJcIH/83TnAFZf8nV/qXZPjrGQSOolefCODDnkVqeKDPAdhwlsfZI375Oih4Eqm2bDbmFwnAEJ70AEVDy8NNOxtlnrkPBmEmv8v/+9lef9CUZqoP/N73bITOVWfeSKGkrIY5r32V60rJXGAJUsNQqL1HJncwiwudmHM5qsGPjVx79E9RecUoUwngt4ZeD4DEGQB1jSbcJgB7ITgRmL2FLdl6R9L67g9s6tapXeteehGQ1oUXFTL69seHUV3FWFTEeCIexM4BfH9ckH3S+lF95YXLionyhDDIWwBWUPzAGpxSaG9P0PYPQBSBFM8b14rDwKv+rt8KCNcGHrPufmlQxVQBVQBVUAVUAVujwLX/lZ1e46rR1EFVAFVQBVQBVSBNgUIQs5cKkjpCrxqvdkAFPneK2n51IcTt9R1ru1QN3zKMjgCCxtete9wGmu87wBLqDbnB3c6fa6WqrXOjBCmAIgVQJe/lQ7uM40yvH/1tXF5+2QOsKMVIl6t1+T85bL8+E0XXFgRAKzrz8gytvZB9xL+FTRFNPCE+UtsXEip6QRarmyufZ6lnhOKMZycsCqFMlU6x+g6K+KRbikvwA2P4wHlYfc9Gzhxmy5ALJbP5QpVsy/Dz8Po9kdgxUHcxvl5TxQAs46fK8nLb2clj+ysoW4fYF4CjqcA+X4v+gAAQABJREFUjje/g9lr4R8s7WP+F9fHQVhFVxfzrPzzOVML91j+OwJhnhfXaAND7kF9l1tHPOyUCrBZFNfNwlq4PWHaUM8qbo7ll6bvqgKqgCqgCqgCqsBNKqAA6yaF091UAVVAFVAFVIG1VIBlSDVkC7Fka8HAB3A6Qc6MlG8bwDKuE+8Sjhk6YfzuK+VVC9a5Sb4hnCN8WjzouJEVMgruz9K5V47l5Oxo0Vwze8oajF0MHx9P1wzwWW5SgplSpRUmTrcQy+ZYqndmtCKzKctAq34AIJYlEj4xl4plbKsdhKOEYRnkO83OVTF/WTLZGo5dRxYUIRFKG+MIO08QUiG4Hceq1a8CM8IlAi1++X3Ik7pWPgOF6NL72ft5+eNvT0u+AvKGDc+iHPbUpbL8y9/eIVFkey03onBKld0Mjm/1ZeTPhJ1Ztdx+S73HbolLDZYGLjd47/fBbZWCPgRghGodKPtcDnotN5++pwqoAqqAKqAKqAJrp8Dyv0ms3XF0JlVAFVAFVAFVQBVYRoEwMnkO7QrL+FSqZWmZ39Y4YBJucYBbpPGh2i5rojOFXevWY7B87NEjYbk0UUYp1dVSuxgAw/4d/gWOlvU4/nrO2e5iIoghOGIJIWFJkNanFYwynDl0Ck2lLbiWMAl3IxfBFx9aoedNk3e13HSEIomI5wos+dnxvPzhc5Mo52uIDxd3F0LBHzgUkkfR3W6ox2fWuVqARbdZCWCMzQFGJysyDYCVZ24XANMc7ieeS2fCY+BUFdCKAJXupaUgFc+FJXgsGWwfdEvxXuR+f/FSWjIllCMCFNG95IA+IxMl+bPnU/Ibz3S377bkc7qu/GuQhe42PxzXQqzFrrelFkGI1ZNcWC641Hb6miqgCqgCqoAqoArcXgUUYN1evfVoqoAqoAqoAqrAdRX4x3+rX86OFOTCZdhy8OEf0VdwxXhlLwLBB7o8pkyLOxMUzCHHKAlniF3Gdd1Jb/KNu/cHpdrokBdezxjI0wHI8dBdIRnu8+GYKwM9N3nodd2NAIsQiO6jb76YkudfmxMGfAcBBJ9+JCqPHSaCWn7QxdNAFhTL6BIRl2QJdLAb96QyLoTvP3p3WAZXUHZGQElg8mOU3P0f/3nyyoFLIEnvoxSvgnVGAI2YRcXtVjvovMqhlM+q1mUKrq5zuLemZpFtBncR5yNEoyvt8lQVpXoo3XM2hNinHfS1H7MVWM5SPzrZWiHthFocs+jIl5nP0rpigMI2NZCss3CVsWzR3rZ9zvV4zpJIXme6yezB+/ZmwvEJAVnt6QWoW+ng+acRlE9A6IE8iajHlGmudH/dThVQBVQBVUAVUAWuVUAB1rWa6CuqgCqgCqgCqsAdUcCNwOuv/d4B+f+em5CfHStKAKVd+4dD8tDhAGDVQkcIS95OXigALokMdqP8q2Ph+2txAg8cDMl9AFl07CDS2wRp29lXhBd04rA8bHauIix1O7gzeF3wsRbrWas5WAr3V+iq+O2XAa8qaKeHcytUqvKfvleVQt4jwzuQCbUMLKKLhw6fHXBFTeKc63WHTKJksIq8Mi9oxTOPx+WXP9m94pywMgDJv/+L6SVP7+K4JScuVuTJB9Adz7X6X9vs7KvRCUveOV1EtllF0ih/rOCYLNTzwvE0ibyzbpQQulwNGepClhVWwk6DruuASpY++heVmGYAyd4/X8KMTkEMmCF5JvLKMLcmyvDo3GrcNoBFMXmdyxUCuqYBdX6fY8XXhPsTXI1NVcxjFdA4BJfkLgT0L3dvcD/+bHA/uvvskS818XPqWXUQvb2/PqoCqoAqoAqoAqoA/gebiqAKqAKqgCqgCqgCG0uBX/t0r3zlk004rloB3jmAovbOdCxf+4uXUnJ+vCT8YM2g70O7Q/KlpzokibyktRx06DCbaPFIwV3yTZSLvXuyAMBVg8tGZGdfQH7ny/0SCa3cqbJ43tv1/Y/ezEI3kpZWyR8MVVJu1uRHbzfksaNluXd/4LpL8cOJQ+h0cFdAypgiCrCRRxliEA6mj30oKffsDa7KpTY9Z0kF+y81qGs234IoDE9f7WDp6SwA1eUZlA4ihL0MXkdmly02TbdAJKyZKUcmauILhAGwHMYxxTyuUGBlji86vD4wbrGa7Nvpl/S7uB8A8wiwePxk1CeP3w9N8ALzw/CWAUp0n633uNmSREKo0UlL+LNHtxlHJcN8Llz3HcFl3XDcpx1ecd8qfphnM2K6GfJ7HaqAKqAKqAKqgCqwegWu/Y109XPoHqqAKqAKqAKqgCqwxgoQHNkVSx6UQ7UDrB+/m5XToygvY1c5lJjRjfXT41k4X+rymSc7ZffACtPIb3LNLGE8BjfPOx/kkd9UNR/wub5jeP7Pf78uv/Orfaa8sZVDdJMHWcfdmCNWZo0XKAXBVXtwPriUTM5aOC+POYellkFjUhyB5Mx+euJ+t3Gp0a2UgMsoeBOQqdl0igcGOlT5XTMY5RQNeQAJUbbIUCl4nFYzmg0H5m4BN7/PJQEcYwbAjKVt7cMCpHnxtbz0RvxycG8DpZEL37e3pVYMa7edeHz98jTgGKBNDhCObkAL8I8lg6y764555PMfT6Abo8+U0qUydePu4n4EgREE02/EUcSNYOHnyYZX9hpLpYY51wTKd683qOVSgz83OlQBVUAVUAVUAVXg5hW4/t++Nz+n7qkKqAKqgCqgCqgCa6hAyO8ykIqlUByjExV834D7Ch+yASKqABsNuIneO1cUfyAN2BWVfUP+VbmAVrNcgpQxBLxninT0sGMd3GJVABjQoDMAa995JY0MqKjJyyKk2GiD8GXPoF8uXS4ZGNO+vm5AKGZB5VDuloy2v7PwOSEW85T4lbhBd72Fe177HXO0hvsDxsW0+F2G5u9A7lg2X5cLuO5DKNUM4JgrHXQ8MSstimNwuJxNKaDsdKnRBJ167idleeReC266hUnqdCRlC/UrmVLMyDLwCa9fGK8iqL0MVxWs/dB2N7TdiVI7BtTvGvBJT4cPwM9lGhAQts4gSL6E+9YJEQe6vNLbufblr0ud32peIyZsB5tX9sWaCUCXGwSF+Gm4ZhMnBCIQs7PHrtlAX1AFVAFVQBVQBVSBZRVQgLWsPPqmKqAKqAKqgCpw5xUgLKHjh4HU/OwcR7bPGGABw8QJr+qAAoRb/HD8GpxYrx7LyT4Ev/+9L3QDdriMW4ZzrNUwQejIQOIHfK6BJWFVOJrqoBx0n5y6UEL+kNuEaO/sXwhC1moNtzrPb6Ij3vnLlpxGjlgT4I0B3yao/rBLfMiFQhzZkoN5YBWU1/EfhoSzrO9WSuF43XhdHz4URhlhTUYQps7KRpbb7ejzysN3RUx+VAh5aASH03M1Ge5ducOOeV0enMzOXp9cGCtJGcdqLgNgsoWGfBPZYLsHfdLX5b+iAcvi2gPRCS2nkfvFe2A6xRJH2sdwLByPzjR+feT+iPR0+pCthfI7uI9yAGBjcGsxUN4el9AZkdlUvL830mBYvQ/3eHvHRYKrMMoqcWmWHSy9ZFg9Q+s5mCc2hy6QRbjoGOweDjhNZhyPoUMVUAVUAVVAFVAFVq7AxvptYeXr1i1VAVVAFVAFVIFtpwCBCccDh6MmPL3ZxAdkAC3ChAaCxEemalJCHhX5xAcXivLCa2n53b81IPchHysIkHUzHdiWEtmNgPNH747IS2/mACVK5gM614B/xQlexQ/9J88XsatDNirAiiMU/1/+18PyjZdm5e0TeemI++XQLrc4GwWJoawtjFyrxeP8WBluo4qBW4NwQoVDLZcO3UU3O5iVBNlk16AX8yVlMlWVFCCVD6Yk5jd1Jz0Sw/wEghxlA4rM0xX9EYJbK51Fd0EApaOAZKdxDqdHAJEy1zqE7AkLxRo6NKbl157pMWV+fH2p8rdcviYVXPRYxInMJxcgVd2UQbJbX0fMK12dXpQv2rSnKUWsvR1etY7XlBS6F240gEXgSxcanY50v7U6TTYkgHMrAESFgk1cI/vcbOVaj4ShA11wzQEGlis1ZF8RPKPcl+AOsvOalwB6D+8K3RL8XHhU/U4VUAVUAVVAFdj6CijA2vrXWM9QFVAFVAFVYIspwM6AuXxCvvPjjIwgyJ1OkTKsMEXAq1ZAdqsTWg4OkH/9tQn53/7bXaYzGvOUCDLWYgzjw/1vwMX0+89OyhQ621XxwdwPwEanUAH5QZOzVXG7UWYIyGHDl7U47lrOEYBL5ktPd8unHgM4wnrL6ERYzJekM+EBwFqo07tnSvKTd3Nwu9WMo+0C3Fv3HQxKb4cXkAOOrWW6Fi63ZpbmBVBy5sHF6Uq4TRfHzrjbhMRXUWYXR0kjy+zopOLwXgeaXO8YvN5RADm689xur3HkkTT+0XdTS+6CJopwozUR+m7JqYsluWcfKN0Sw0BTojecAMHqALKv6E5DbJRw/UPo0OibB67cnetg1tfiwW1a2V6L37nz31Prw7uDMjplyST0YCA9r0MqZ+HMm+ac27PA2ldMV14cYE8iXmSE5U3pJt+ni4tNGAh544CSvYB8dDTqUAVUAVVAFVAFVIEbK6AA68Ya6RaqgCqgCqgCqsCGU+CJB2LCr2Mn8/LmyVbuVLNRNY4O++MwH1n6dWoEXfXQGc9CuZpv5dVnNzzno4dDCIzfAZA2CzdWBs6S+aBxQI2yVTdg69xYRe4/sLF/3SBgiwTdUig6ZGbGBQCzMJMpAyfNmydyKB1k18JWyR+70b17pmkAFh1UNztY3udx1808J84XUAqKEr8myhkRfs4gf2ZG2fPzenYsWttKjkv3XSyMzCnwlBzOpQfQ7Z79fnnnNMKw7MnnJ+pHF0JmNLkBskq4huy2SCcYYQ7Biz0Ibghe+JiAmy2dsXAMFzx3DgMsezoXXnOW3XUn3ci/chpXE/OvOKfP65BAYOG29jFu1yNhHLPCrjdYFhsC0GRXwtMoj03laqZE8K99LAmH4dUyy6X2J5yjI5KDjwWEwPN4lyeqcuZSVYZ6XfKJh2MAi3dWg9YK9U9VQBVQBVQBVWBjK6B/W27s66OrUwVUAVVAFVAFllXgbjhkupNeA7LGplpAgp+XWQIFlsBGcCYjy0xy/c/oyx5juTcTMZf8wsc6JJ1ryInzeYTJN8QJNxEdNwQn49MVOQIXi11Kxg/0BYCuCuAIt2EeEEuu7vQg3OF6PEv8ZnR5sgznTAOwBllOyL+yACBIfhqNGsr9qoBNS+y0whNiWSehVT8cTNOYawalfQRYSZQNRkIoV0OYfL2G7xHETsdOEHqtdgT97GLpMhlWLiw1EnLLh++NI2Q9h+tTMw4jF6BVApAmEW2iTI7AyyNdca/JVvOjLDSCkkoyHjsHi5CL9x3L/7zuBvLDfKYjJt1kgz0e4ygjPGXJne1OYznjHoS7s8sj3UwcdG91w3l2JwbPpz2Ynh09G63YqivLYekkyzYnkff14htZOT9aANhrlY6eAMz6B7/Wh7yw4JXtFz8hhGQ5ajoLhyS6NLIc8fnXUugiapksLULA517Jyj/7uzskGVv9tV18PP1eFVAFVAFVQBXYygrcmd8YtrKiem6qgCqgCqgCqsBtVIAfkPtQYvbbv9Inf/v3ilIENLBxEIFLf7f7Sg4VAc16DAbFH9oVkCmACWYGCQ7DD+bsUucHDKkgONyD2jQGfo8ixPvSZeQwXSqZtSYSXvm5B2PSvQE70VErQg7TVQ6Qja6k4rxTh6VgdNQ08fosIE4Xyg5vZpDdEfTR6dTT4TE5Ze08LwwNEygj7EOpWfvrqzkWASHdUQRIdJuNIjg9ASDWk3ShNK6GcrYGytzY2RJln86auZ/uQ14WoZUNmnhsztFooBQQB6ejiuddQPB8HplOLLEjvEoi+2oc5Zhed9WASTduwh7keJmOhdgvCdjH8ky69RjuzjXxHr4dgzCq1cmzae7PdBYZXrw34YKju6yG51W4reYvsSl/5fboT2AA8ZmRgnGw8dpTl8nZivzBN6blH/3WsHGSLXUOvEf6ce0IbKfTVeSPleWDixVTNsnTZvfOs4Bi//sfjco//TvDS02hr6kCqoAqoAqoAqrAvAIKsPRWUAVUAVVAFVAFtoACu/r88s///i75F/9uVHIAC4Qtfeg898tPJSUBaEEXyGrzk1Yjy2EArNGpCqCGZXK46KzqhIOHsIKAgGMO7pOxyZK8fqJgPtDztf+fvTcPritNz/teAHcFLnaAIAjuO9n7Ot09e8/ikTQeSyMrsmx5qTiWHZUTK/FSlT/icvJPXOXEZVclcVWWcqzYVRlZiiQr8mj2rWd6unt6b3az2dw3EMSOe3H3C+T5fQeneYkGiIVkD5t8PzVwgXvPPef7nnMuB+en533ewpWSut4t2lef7RPc0KSXDRxbdLADfuH6uZ1rWHbo8CvgpkvgZlt/2s6HvLHW0GlxYaFFnR6z2ibqLBcC9gUrKJ0ESLHutZxlgJIZlaNRUsZI6aTVlLHVvMa0wF9WkAdocrNZSTihAEi4uKZmGzYx1abSRXXW01wTOkeXxop2dWLW9uzosi2CaZT5AaWaRwybyHCaVNh8SY6010/kVWI5b7PqtJfQGh44kLUvPdOrMjucX8Cxqg3UE6EcUjFjwckGTOUr3l/zMW7Hz+RzleSAYpAHVtS863qclv5f/8lMgEst1hBsa7W/vy1qhMC5AUDhkpsT5JstUP4XWbQSwsS48i5pbZfHy7ZnhGshGlyrDLoQlnUMBo62kUGzP3t+OjgiAWDsG3S3II3eFdTiOrjZcxwO5t9cAVfAFXAFXIG7VAEHWHfpifVluQKugCvgCtx7CjyucPd//z8cUJlTSWVOdEITrBC4ag7Tvl2qkKv00MEOHTsRbuyBTTh4RgjzVs4Rg5D5C2O19+FVPJeCarKOKf/pkw93xU+FR27oZwW9mkO+s8pjAsYBIUpy/vzZT/OCZmU7JJj0zINdgjPXA5frdrjJX3AJPaLAdrKgzgm4VVVCtl1wcM+2KFCMsrPXlEUGsCIAHb0H5SyjAyMAbLUB4IjhFdvgtJopCHzBswQ4CMUfUHkd+1gtLHy1fd/o+YzAUU8nQPBaIBpAqleh44qLD6HvdUHDdCcZVyvviXI4uuq9o26TP3hpViVy0XlqLNa0BnU5lHPpV1VayvrICwPA5ecbdk7n/+SFYig37O5M2qcf6bTHjmoyt3HgoCoLWMUD3fktrzyqP/z+ZHCfNTRP3IOXxhr2j//VefuNLw2G0kc6BtIkIZtOmBoJLp1jleVqB7incGhNzS4KYFGOWLdzlytypQl06fkOXYvDg8lw7oCSgEI6F8auNubDPMLMtHPYGODZhyvgCrgCroAr4AqsrIADrJV18WddAVfAFXAFXIGPpAI4WvbtXD2T53YtCsjyoILiBxUyPqG8oIQq6ob60spuunZHThlhSXlCyweQpCSYsHwAGprhFa+XVGoHkLg8VbV/8W8vy0VUCXDkey/O2ndenLN/+NdGgsto+b5u5nfAVWtrUhCr1XaPpAOIISuL4G/yvCilu6ISvDYFadUEdfKKIuvKlqwuOBO5tFY+OiVtzQMNt6kcFAcP7qQYHuEAuhEIa97Hen8GzMQDPSmlg6RklZdFeSED51su3mjZY4uuM+Z/4lw5glfSIi69q0uDt9W18fNPKg+tRfuVzwj30xWVGb58jDB8lbmqxLQg4POH31epof7v4cMdt82NBUSLQ9pZa6z6qEpexyZqglByAsphVVcLRdZw8nzR/sMPp0KXQbLHirrmssoBw1mX13Y1wTjew57ycmU999qMIF1D62kEWJcUeOQk4vrivNGhEaDLOLovZ29Im4ZgGc9QRtii6+bArnY578Im/s0VcAVcAVfAFXAFVlHAAdYqwvjTroAr4Aq4Aq6AK7AxBXAgAWD4WmngMNqiskJARjwoU8sIbK2UISUm8IEB3CEc/A++NW4TyiDC2QN0qGjjd07P2+/+6YT95S8NqOxP3e5w/wCJ9B8OGBwxmwVBvD8laoX767JyvHCTlbWMpAhEvqCfBTkAIoTWtwjatApK4ET68qcHbP920Y8VRpsgDvNvHrijAGbADdYFyGouKWze9mZ+jmETjwSLj6oUrignXJvK6Pr6ATatAR4CflYqhexSwDzvLQra6AyEn5k0hi2er4kU4epqV5M+MtIoXzyn7K2q4FZN50XLUn6YnEza7s1TpRBi392ZCKDnVpfRNbvXWmVvC9pqnq3or8kCVYFXwC3cb4T0T81JC22Y1rxxTOGse/y+Dnv1eFGB7jrHS+JPqGT29749pdLAvG1X58XH7+8ITjqC+AnPn9W1sVUZWGhSKC6qFDVhR3ZnQ2fQBWmBu2vncNr+6i8NrKjzzZxjf68r4Aq4Aq6AK3C3KeAA6247o74eV8AVcAVcAVfgDlUgJRjw6ce6bUquldGJigLDF61dGUs7tmXssG7qlw8cSHGeUPwawAfXyqWrtShEXWAAl5YeBAkWFZJdtPfOFYP7ZV6uLpxgw8qvwgVDuVfsLor3t9FHYAiB6rP5VpuYqUbHFfiYK+LCKgeHGHCjpbVh5wVsXnpz1nZvHQgZU8uPlUnjYrr+WaDWejozoktR4IVHdMIdtBE4hyMIPYCBM3mV/MmBpV3ZgoDfROismAylhCvBK2ZMmeTubVnbKUfaWyojXFyM3EZ0c8RR1dOVVJki5Z5J6+4Q5NO5iALRdaaW6E/QSfsC8kypBJG1czzWD8S7VYOyPEAgZafsH6CGI+zwzpRyx3QcXTcAJuaDBngG04iqbcvSqF1zASz2jiQFLVVq+B5wrm4z83ofb9C7JqoVm8tXrCgdP/MYAfiJADjbM5HT6orgJsenIcAXn+qx+3ZXbVYOtJ1bU/bo0a4bOvU4gg9XwBVwBVwBV8AVUKm9i+AKuAKugCvgCrgCrsCHpQAB4n/pC/125nJJ0KJuPeqyt2NrJgSVL59DCM+epysceCoahIoDsdrl2lqUKyb6v+i1BW2XFrC6rLKwghxSuIhoijimMsOKfsBp1Na2aPt3dCi/Kh1AWLzfjTwCRHrkJpsv05Vv0bYOpu3UpYqVBX/IPoKEiM2pVG7Bzl6phrDzIUGv5QMI1JVrkfNJIErbA5U6pM9q0Ch+P9CK8PHm8kq66AHn1guxcJIBjcqCYOiJrm1tyhVTaRwwC+ca29xoDCqf6yuf7FduVMWOyf22IOcSgK+rM2V/4dO9dlBlcWQ+RXlZi7Z9KGUXr1SkT6A+6kxJR8MWGxDsoiMhg3NN6R2us/UGvFcFhl45XrAJlXJm5Wh64EDGtqh8tXl06TorCZCRzQWM6utWwLx+/rQ6YP7pj6a0XjUf0LQoDWWedJfMybHXuQTTODe9XSm7oIwryifRO4JX144iQ5Vd1HU9OpKx7v1yk+l47Rky4aLrokUXRVbQcmGxVUH5XH/Z0PFx1/DmO0xeO7r/5Aq4Aq6AK+AK3P0KOMC6+8+xr9AVcAVcAVfAFbijFABW7N/xQcfV8knS2a9bgIfcKxgW4AioVVTu1Cfkcrn4DTleGpGFCfdMWhs8rCB7OgGyvf4LY3yqYV//8XRwKuG4+Ul73j7+YI89+2RXKG1b2mxDD4Ca/u6k5tJQWP6iDSms+8Q54ETkEEtqkYRyA4jGZtTdbmDl3QN40lCTDYy83D+zAliAEeZBl0IGHe+AdusZlOn1qrvifCmaZ5vCrgBuEzWVdArwRN0jUfXGY5eC7H/nr4zYD1+dteNn5aoTnHrm4R57VKH3ObmQOGfAKADZpx5OBDj2zumSsBUgKRHO78G9AphN8+bcUf6Z0toYgLrVYBbA60+fm7YrKicFVjJOXSoLrPUJLF4fKoVOy6+6Tz3SFWDk868JgE2XrSPTsJ7ejEoJW3VeU+E1HT44zkYGksHJRSaWLrEPDOChpm3T81XNNxuC/B8+RB5ddA2PT1M6G4Exzh1ONTK2VlvbBw7gT7gCroAr4Aq4Ave4Ahv7i+keF8uX7wq4Aq6AK+AKuAIfrgI4X5JQkKbRkW2xX/h4X3DAfOP5uVAC1xM62uVscCCtrnACBQIJVIEBOF49UQguJ4AKji3Cw597dcb27cravpEPOqOaDnXDH3FLAVdaWpK2T2WQ54ZKdnY0KsEjF6tVf2VVNJVzl5TxNNBmA8r/utkB+BibrIcSNpxaKXV4BPZ05eT0gZ5sYCQFlih1q8lNJESokk7glQCLnmsGSjfaJfCFzLO/+LnBoAWnCpDT7CIDRBJQ3qncrL/6i0N2Xq6003Ks5XQeORZZUcuPx36jDn9aq2xh7K9bwA1o2DxOnK/YxavK1hK843y36XqpVOv2kzcL9tVne5s3XfHn/p6EbVd56S99ssfOj87b5bGC7RjO2O7t7aF0ELcbOVk4p/KVFhsXjAxX4xIsi3cKamPdKcG/XSqtfGB/zrbIoQa4Yx+A11aFaTWWbFuAR8o+gXttvNGHK+AKuAKugCvgCqypgAOsNSXyDVwBV8AVcAVcAVfgTlMAl9BXPzdgv/JsvzKUlOOkr5pK34I7RvaYhcxiKPMis6gox1JrACsCNQsRLMgLihw/O2+7tiblEto8QOhUaVpO3OjpBzpDqdwbJ8dUMriUTSWXzunLRfvmC2021E+mVDIAi5vRknwqujOGroHaEYCsQ6V+88qyYlAiBxwBrq20LtxslPThcuJ1tqvV21ReGdrqBQgEDFurfDAcrOlbYIzr1JHcJ74YzIew9+YB8CLcf3SqJqdZVa6tyIVFlz/OH6WLDNZwaUyB+ksdFXFgtQnwkXfF+/TjmmWiPQqOHxH0nNMcOrMLdnB7w3bv6A2aUkKIznmVsRI2/9O38jY+VbL+HspXza5OSTMdg5HQlIBRB3am7eiebIBXXQJuDHLBKO+kqyPnj3lntC3rYK0E3PtwBVwBV8AVcAVcgbUVcIC1tka+hSvgCrgCroAr4ArcoQrgzAFC8IVjB+DQIwBzWq6nTCoKie/MybWzIEiyzOnSLtBB0DvOnpsZ7JYA8gvKd5qYrQRAwXOAC2DFCYXKn7pQsoMKqqeT4WYHGU3FUrTGq9N1wauFAGs6BaTS6bqNtJGrpJJAZYOR70ROFzArHnRKbIZFUah6iw31Ja1XDraCQsnzcgwNCtA0O6ji99+OR5xXzJDyR7QC6ODIYm15OeXyQKslSFQRyMJtRTdL4Bsgrkclps0DjcLaO6N8MjoNrjWGVBrYPt+qtet4CucHCAKvGIUi8xL41Bxm8wTVk82lklHcVXKrjU/WTFILWCVt/86MPfNglx3e0xGuB5xxsf4A153K1urT3MkXY/5JlbMCEH24Aq6AK+AKuAKuwPoU2PxfUevbv2/lCrgCroAr4Aq4Aq7Ah6JAXG4IfNi+JR2cSgAHXEk/fWNOcOua02dAOUcHBRxqITjp5gAWi3v+zTl76e1C4GT8DnMBpkBnqjruaydLcmDN2YHd7XZghzKW1uYq7CYMwA7wieyrqzN1BaFXbWKuZsdOlu3MpXlrE/T58tO9IU8JKNUtgEdJYLmyKEBy7UC4nZYPSiBDsDkZVItyHJWiToDLt7udv9NxcHnXQUo943D75mNzvgBXhLIDiHYMp0OXyVFlYMUjpXrFhw/nQkljVesDit1oAJkAoCnllimR6n14hTZ8MdgGsNYjMDaTN6vp5FJWuFXh/AN9KXvoQNYApUBD8r0AcMEdp9AtygcZlEV2dfif3kEM/+YKuAKugCvgCmxCAf9f0U2I5m9xBVwBV8AVcAVcgTtbAYLi4xKuLzzVE0rSXn1vzhq1VhvsS9gvPNMb8opWKrPb6MomFc79k9fyglLkYUVGL6KOwpfYBeVwnYIYBLyfVZe6RYUiHRLIWs8A0oxP12xOTqQxuX1+9nberk5W7eUTRZuZjYFcw/63P7pqzx8r2n/6lUF191OukmAJEKWxkAgB5LjBatrXNZzVdPSI0TQ98eH8iFvuuddm7UfSbqHRao8dzdoXnuwNwGe1LoR0bowHmVfs48n7O+2t04JvWm+7uhAe2ZsN2VOUljJaW+WyE/CiVA8gtd7QdLbjfFL2N6FzwPEGe9LSt67Sx4acXupaKHPfE0c7batKRGkuwPw4LkAOp9V8uWZblLPVvuToiufuj66AK+AKuAKugCuwcQVa5ubmfk5/tmx8sh+ld1SrVbtw4cJHacrXzbVWq+mPsuuDUq/b4C7/xdd/b59/Pr+p1M2HLX8UPyaUytR1V3avfv59/Xfv+cfFNJFfUE6WIINARgSaCDFvUTlX9GndzPnHZfPayYa9+HZVsGJBjihTHpXcV0v/ALDv4UGzzzycDKHlHLervdUO724V7FgRJ133T8ecyvom5xbkplL53MSCgsYb9u6Fhp0fu26z93/55AOtdt/eNhvZ0hbWhhuoqPe2p+nsR5YUjqBo/bwJh1p3B2VxvLYQvhIEOt3mQa7V7/+golwplUJqfvwx2qbDPryvzX7jC+ngHBsdp4PjQgCA6IbDrrfTFOQeaTelP2FfP1m30cmGAB/dH1ts3/Y22zvcJoAETLqmL+tmrcGFpsecSvd61Hmxeay0/u+9XLdX3qNcs6F/F+UUU9lgf7cyr7Rr5ozDbUQB/YyCyjvRNq0/n3K6rvLzuPAWw3lol1trqE/ZXtf4W3jPnfRtpfXfSfO73XO519ffoG2qRhv0/x4cvn4//1GJ+B38j/QNPpeZTMZGRkZusMXd89Lt/wvl7tFqQyvh5nffvn0bes+dtPGpU6c+0vO/WS19/X7+P8qf35u5/rmBP3PmjO3du/dmdvORfS9/wPL/fNi9e/dHdg03M3Hg/ejoqO3cufNmdnPHvne/ZkYpHiVzMtcEl0yzAwt4ffXqVdu+ffu61gCceO9sWZ0PyzY0WJBTqmqH9y0oWLyhcj/lJekg+7dn7OMP50LpGcfiD2TK33bsbA8dANc60Oh41bJdKh1UoPmg1axQLtnciRm9beX//+OZK0nbtStnjbYOG1Z5XYBW+lzjyKK0MMwrFLktBpjToU54ZEpRQlcul21ycvJD+SP45Xfm7e1zF622EHVrjHRosROXWm28OGBDw+12qK/FBtRRsqIgfkLl6ehHWWFajwnlUJWUN1ZdFCVqrVlLIsrKGptN2Da9d/sWlfSpwyFOLsoOT5wrhf1w3ju1r3aVAnaqKyTZVfEoFos2Ozsr3YbDU6/I7fbW+QmrNuTcUnFhg+R2ZYyNDHfaTnUuTAtKFZWhRo4Xrre5wkLotBiuKW3a3qnrTFp3qPsibryOnmTo2Bgf7057LBQKNj8/b0NDQ3fa1D6U+ej/qS+YWrHBQRHne3Bw7fP/wOrv778HV282PT0d8u76+vruyfXzbz/wsqen555c//j4uP5NT1tXV9c9uf6P0qIdYH2UzpbP1RVwBVwBV8AVcAVuSoEob+mmdqGOdIs2JqA0k6/ZudFKcHVtV0A3mVfTyqbaM4K7J2dP3d9uKQGissrN4kGJWTopGKNSt/UMwuGjobJAAJgsPG2CYKsBrLomAaTRIeTcqoewcABL5ECS+0vwbFQliBXlYxUrdXXVU7e/CWU59ads28B6ZnRrtpkUmAqB8tdxuCh8/tipQujkx5FYalUAa1JzLQuw1kTkAFLTcqVduFqWbUzv0Rdrpj6yKK3PX6kpmyptqcSC3t9qx8+V7fJYNaw9JQfVjDoucthUshFyw5ohJsdkcC5/+mZR0EtWMe0X7XHx0fHw8tWqPXqoI3QeBG3Nq6OlTkzoNAgw4//QFvBW1Xwn5BSjG2Z2rKK55mzHlmQAmdGR/Lsr4Aq4Aq6AK+AKrFcBB1jrVcq3cwVcAVfAFXAFXAFXQApckisKkIG7CfhBJ0PcQUeVvTRbSAa3zcce6LIje7LqSli3Y6fmBb0ogaN8r812b0uHPKb1iEmO15yAC0Hg8yFgvdX27cja5KycRyuMwX65i9IJeEoALlVlRAGwyG9izMvBhFOoXK7bmctVG5+pBrhFhtMnHswoqyva7nZ/p/Mj4eqUYDaPVuWIdbVfg4EFzZcxJndbiZ/lgGprW7SGxJ+eFdASX8rICRUPlomTdFbwDvdTQ9sAGgFSZWmxsEhgPOV9CzbQQ0i73rACSyxKn1p4UfuGXC0N9oPzjn3QLABIxnnBFTbQrQh4TYB8rGIo8Vy0KYXuvyv313ShFsAleWZH97Xbo4c73z8n8b790RVwBVwBV8AVcAVurIADrBvr46+6Aq6AK+AKuAKugCvwvgIVZWiVlHXFAF4BpOhIV62rRE+laf3d6krXK5gleEW5ICVq/Y9021Re8EUghDI+XDrrHZlUq40MZQRIRGo0Jmca2k1GweINO31RDqSmMaCwcLov9qkkkPynrN5L5hMgJw4RL4fueHV7+0xZ7rGyjcuNlReAI7D89RNJ+/VnkyohbNrpbfpx70jW7t/bbi8eqwsIRXoyh20q/Xv8vpwC0BeivKql46u600pyMS0sNgToAEat0jIp95gAnNxkMKhF6U2+2agcUrjMOgWYyDjLpPTnrsBYqQKM0jZyYdHlEIC2atyPttsxlLBzl+l2KMeWGBYB8C0CYAe2p0JJZlr68kX5ZSjRVFki10S9kRBgbAgO1pXRRWmpShw1vwXlcp2+VAlZWoO6LnYPf0i08DadQ9+tK+AKuAKugCvwYSvgAOvDVtyP5wq4Aq6AK+AKuAI3pYA4giASzpwIRqxUAnZTB1jlzXQEpOyNbndtAiW4hTpCdzk5gsRgtg2mg1NqSG4m4FU8KD8bFFza7MgK2OxQiSJfB3dm7ZV3CjpW0l57t6hsp6LgSqttGUzZdoWJ7xrOqoQxG0LKCR0f6IlgD2WPhJznsknBtlrI66KMD2cQcKZN1qGxybL9/ner9uDhhnUJxtzOAcD5u39p2P7d19vslbdZgwLYd6TsN3+x37b0p+3yeCXMK55Dh4LvCWrHXbW42BaA2+6tNNtYDOWEOMxwa41eVdmfYGI6qZK9jPKwVBrZq0iXrX3J0IWxXGtYh/TqUSklcwA8rTQIjH/sSFeAUGcuKhtNDjB6OB7YlbUdW7MhPy1+H6c6p/nF1yGPW5R39fq783KNCa5pAzogAuiqOv74VF0OsMhZFu/DH10BV8AVcAVcAVdgbQU2/9fU2vv2LVwBV8AVcAVcAVfAFbilCuAgOn+lYgUBgLrCtTPqrrdra9q65YK51QOwM61StFmOpeB3wFlFjqbRiVpwCAFVOjsS4dgpEZThAZUPCsQQ7A0AIW/rVg9KCh841G6J04TBJ+2xo50qYzPBGoLNFXremwhghuPzheOriaXZjNbTIeOPliINFwIkImC8yuI0JuQc+oPvTtpXP9cX3huevE3fAEi/85dHdB6VY6UvcrvalmodC8XEUobUgmBUa8gMm5aLjTwx4CEjk03Ypx7NhPLMK5MV+8Er8wqpZ19sF61PEVWW0PsBSlvkjKvVWmz7UFbdCiPYuNrSOHfkhX360R7ryhYUnp+1IWlL1hkwCpCJOw5tceHhxGoeuLv6utssp32g9vvnAJkl+CrcrHkX/rMr4Aq4Aq6AK+AKLFPg1v+1t+wA/qsr4Aq4Aq6AK+AKuAK3SoHzo3LDKEB9YqZm0/m61VVWdvxs0R7Y12nDg4kAEsh82uwgkwk4AYh6/UTBvv/irP3orbyV9dygAMiTyrYaVPe6tMw/tXprCHDHcbVfuVSFYuRm4thzgkPUjfV2JgPoIBvpVo3+LoGrI2RjRYHmBIQvCkB1yjW1RU6vxgLdDqMSx+XHzCpLa1BupBGVGr53thRcWQ119GvV+3GKJcRbSpVa6Nr3uODY++BlaUdAvXG0V3lcm/6KpEQyJXgDzNnswLEUu5fifVAOyXOzOscNDqpmgY8fydmVSXViLNYtoxLJgV5gkgXQRmllvYZLSrBIU8FtRobVjODjsMoR+3vabEjOLjoQHtiZlhvqxvOlhJGsq5SOO6KOg5zLVEI5ZAJoCTm8coJW7AtwtVwj1tCepXw0pWslKT11PWlfCb2f7bsEWwF2r8hB16Hz0afsrFxWkO4mrttYN390BVwBV8AVcAXuZgUcYN3NZ9fX5gq4Aq6AK+AK3EUKULr33sV5e+H1gj1/rKjyLAVjK89o//aMHDcLyp3qUHlXSs6dKJdoo0ufkcOnIPdUu6DHWXUX/F/+n4t28pI6zC2Nc1fqCnCfsk88krMH93co+ygZOs9RPkZpGJyFMa/yRuYThp5Lq+Ngj5xTOHdu1QB2AHg4Dk2/cfxkpAXjRmwGgDIksPKE4NSpCyW7oBwsOveltD/gVlo5TQA5AEte4fHAlniQAfXcqwV7QUAvP18LWV4Eyn/m8W7BHEEY6XCrRuRgagklmriukoI/HQJGh3dnpe9CcMKhJwH6QDdcV63SA9dZjaaAkgKIxTkBzA0PpNSZMBWg01rwijWw38DN5AhD5/cuVuySuggS3t4urXYMp5VNllrRZUekF5fCDjkD84Jt7dJ8CtiqOtMeOeLSWssffncqAMhMpsUePpizTz3WJUaWWHF/t0pT348r4Aq4Aq6AK/BRV+DaXyUf9ZX4/F0BV8AVcAVcAVfgrlbgogDCn/1o2n76VvH9dVbkPnrleEnN6doEk5LBXdQqRxGlhhtxBQGC5gSuZvILVjcd57nZ6+BVfEA6Dx6Xc4kcpB1b1eFPAIIOg5QYMgA/78Mr/U42FrlNgLGeW5wrFZUJUqK2sdHTlbCDAj3/2a+QQTVm5y+XQhc9wtGH+xoCWFG4eMuyOrdjp8v27Ren33eakfNVOlUUlGlV8HpXgDthJpICiFaTFlWF3kNzKMljvhsZgKOEnEwdcic1D1xLfDGAdpy3LXKVdWRTKudbEJBTt0HBNsLrKQN86FAuBOED/XKCYOsZnEcGEOvMaMnOKDAfRxb2LvZ9ZVwuNJVj5lTG2BzKz3UXOfii8753e9aGlcNVEHSjk+SMnGt/9P1Jm5iLQvltzmxKnRLJUnvmoZwDrPWcHN/GFXAFXAFX4J5VwAHWPXvqfeGugCvgCrgC96oC3JRzg45LBUjwURiAAcDRqyeu77wXz/2N4wV7aF82OHQABUCljYyiAMPUbN0uT+iNM2V79b35Vd+eV+leqVQXkIlgCK4m3EANwTTC0ptHHBK+/PnmbT7sn+FSgB2+/ru/vct+8PKMXRAczGV1PVhRweYmt1MiOJ7iuclgZi8fy6sTYNSRj86LOJ3Q4Y33Wu3+/TkFw6vcT8/T+bBEZ0AdCIm4xpIq+RuU84uSvAC1tGN0W8bI4sOt+xGnVn93iz1wIGeX1H3wbe2f7n81lQB2yyH3yUc77dknuoMDjhD19Q4+GzNyTb2nkPwfvlxQV8GqSgjpkignlxxqKVUVTup62TbI+Y72C9iK4RVrrNbJumoJzq/+UEq4YL93bNwmKYvk88dkpHVe5ak/eT1vB3cp26ytEZxszU0A1jtn384VcAVcAVfAFbjbFXCAdbefYV+fK+AKuAKugCvQpADd+772zUl77jXlOlXqtntb1v7hbw5Yd3d701Z33o+TghJvnpgP4ekrzQ7AcnG8Hjq98boMWeseOGsuXa3Y6+8V7fylmjVaCgIMqxMwoAnlaJSTUdpGGDmljDVBC8rV4sFrsTvnRmV98fY8AupwcM2rox6ZSbiI+K9X7q1bWYIYH5N5ffaJHjt9qWJXrs7b7FxLyO3aPaKcqKWlAPcIUCf/qa41VgSHqjjO9J/itkzRX3LB5cO1VNa2ZFABsYA5AXTp9bwywdoVeE/+VLcMXh0KoCdTilB6nFkA1ZffKdrJ8yWF5S+ok2K7Pf1Abl3nEdhDZtYvfKJPeVJJdVOsaw6Ltk/dGB/Y3249XRu4GJaEAfK+8Ebe/vD7MwJZSoTXYoF6ZK9xDAAX35oBMI4z3odel3U9kX3FNrlQXtkWgNXUXCNAPsxcjAVBz1atvyIwSBnklclFO/tW1crSeLDT7GMPdmsXSycieot/dwVcAVfAFXAF7lkFHGDds6feF+4KuAKugCtwrymAQ+Tf/MlV+/qPZ1QOFQGa196t2X/5z6v2f/2T/XesHJcnqqHMal4coUUsYvFaLNX7c+YeH/cPrhmASL+CsdczgFeUcJ0SwMGBVRJAkscolHut9H5QwhNHOuyA3DLbhzLWLRdT3G0QUFKqyIlF2aD2Gz8PgMAVttagFK4k+EFw+cx83X706py9e2Ze61Ju1WDSfuXZAdsvsMR+CTI/fqZic4JEWYG0w7szcgOpJeMmx17td6h3wSYmSrZzR+Z9aAJYAqYpTUrdHlN2VqV086qjrAlgsaIA8JSTBZCZlTMtKdlr0nBCoerzcmcRtH/xqoLX56sCS9qPAN+wyv12b0vbE/d32R49bhtM2XdfmrNj7+WtquMxRicqNjFdsy9/qid0I5yaq9o7p0uCQ4sChgk7JP3p8tfsquKc/+IneqVX1DES99dm4c8r7+Tt6z9RuaRy1oBXAFKAXEbQbWKqoS6GbTairpPNZZEhOF66nFS22Jzyw3Cg4TyrN2p2Xplq2xScv00avnW6TWsSDJRo6MaVSkfJ8emG3HATAX4JA3JY+/YL8/af/9oW6xck9eEKuAKugCvgCtzrCqzvr7t7XSVfvyvgCrgCroArcBcoQFnXc6/m34dX8ZImp8v2R9+dFCwgDvzOGrhZCBMHBGyT62lI7pdRdaJbPvrkwBnZogI4ZU3NJGoB8tAhb62BJmQm1eV4yYj/pCh3k4WIrnDbBnDc1OU2ivZCad0vPNNrX3yqx3YqoLtPWVLNAANgkVOWEV/Mm052ABayl5q3W2lOACFC6ik7o6TsW8/n7cW3ZkPuEnBuak4OqfGK/caXBpXzlbJjJ5UDJrtPUs6vmbkFG1Nnxo/dZ7ZzWJ35mlxgKx1rtefikr9m6MO8GMx/7/Z26VG1l94W0KnXg7tsp4LM98lVxXvqmjtAS9jIqlpLvrgQsqJmBJ8AWvlCLehRleY1batdyOXVFcoNcV7F8Irj4WQ6c7lsb5wohvXgGKzUauHn0YkWdaKs2jMPd9uQYNjygQtu456ra3vBSffqu6UAmDjnprVDk+iGWMWZ17JoT96vXC0BqeaB7uRiAURRLWSANQRTdZ3MCk72CVo9ebTDzl6q2ik1I6DMEa26O1NaS6e9LVh5Uc4tzn9ax2QdZ8fm7f/97rT9ja8Mhs9A8/H8Z1fAFXAFXAFX4F5TwAHWvXbGfb2ugCvgCrgC96wC3IDjMFppXJ6srvT0z/05nCoRQjF79Gi7TRUq9kM5k6amr62jV/DqicMdcrhkBEhwDC3I8VIOWULt6RujDFxpMIm0XEydKnGbEg9Jyy1VX1QouAK/Hz7YKXBSVxla0h5X577tAjaAK6ASgCrOwVouFI6r9biu4vc1h4afPF8UzJAbSU4nHDo4ezh3swJ5r58oqQyPbok4zVqthouHjTTHd/U+SvI61Q2QfKlbMVhnPCjTe/BAl7RKCtTUbKvgEeCwIiBFKSLB6iQ7AX0oncQJNV+uqyxSj3KMiVmpZk5ldlazmYLZ5clKKF1k/zOCW1Cr5hLMkt47JlCVlwOsUKoGqKVNwphWKd7Zy5UAh251aeW8XF4LaiWYFRwsiBMC9lp1XbC2vp60ff6pbjnHoqD7aDbRd8omM9IgKQFqej9Qj9ysGAhyHikn/E++0GNvnWq3MbndaqrB3DeCey4lYFlQppigp3YXX/P8cOlqOeRtbZXjK7gRBVV5nfNBd0UfroAr4Aq4Aq7AvaKAA6x75Uz7Ol0BV8AVcAXueQU65SrqySkjaOr6Grw2BUYd3XdnZmBx8x9AjehHt+DMF5/uUxld1s6P1UOJWpeAQHtHi+G26lUpG+V7OKq4w7+qkrWdW5UjJQCx2sA1g2OqPZPQ+wVKVJaWzAjMyOmVUA3aHjmagEHdKg8EhhF8Hg/Axq0a8b7EPdSxTjxKLqBrA6BCEV9LWHOlmgrQCsiCiykGOASHU0Y5X6J0jk6A1zKaKJWkUyKAiDUDzJj+jbTh+Cn9pch+ywKJzDEt09FDB7JyCWUC5GGbmurrunMJG9A5oMwOCANYYb5ApsUFgIsgkL4v6pjBraV9VpQRRckhnQSz0nhOkCsGWOjQIhCUk97jul5jcAUUww22oH0VVGbZrBJzuRUjIyY0pM6Bk3KblQJ8i8ptuR6O7O6wzz7WveJh0LRdbjugXlFZc/GALwIVuXZmVGaZa0/aUw/gHOsIJaNAZTpVclzWCQyMryzOY1qdLqvS/8wlOdoE/eJBGSldOCnH/DAG55ayTvLohvpVQqkS2hjOfRjH92O4Aq6AK+AKuAIOsPwacAVcAVfAFXAF7hEFuNn8W1/dYv/sdy+GDB6WzXN0cPvUI1220LgebN0JsmQzbdZBOVohKuECwj1yJKdyx6hUjlD6M8oXIjsKgMAjHh9ASFm1f5MKJR9UB7jVBuV9lO7tUQbUe+fqNtDdar19ChpXWdc+PQe8Yr8EqxPUHg/2f6tcTuwTx1IAMyIY/ZpDZ0erXEt6AeDDg6BGQmRj60BabjCVpMmNFZxXvHlpEPROyR/5S42FtjBngEiLSt5IrCIrKy/oUVCgeudSdhfwjp+XDwDShSsE2xcwd6lULxFKFyviJwCagZ5WBbOr2584zaB+p+Mfg9JLgCMgZ4/K6cbk7CPbK51sWDWsJDo3gKtsOmHbBIpycro9eihnP1ZoOmsC4CQEr+4XKOsRtGxvl6trWhlnmhMwiGsW+MYxybm61YOMsacf7AoasxbAWlUGsYcOdunzMyjQhEfqgwPnGXBw+1BaOVbAr0aYZ4/A6o6tmXBNAlmBW6yF7XltTlCL0tFPPdZrZy4o10zOs+ikS08B2r3DGevSOT9+hgvi+kHp6A65AmPwd/2rt+43SiO//tyM/fj1GSvq85LRCXzoQLv9nV/beusO4ntyBVwBV8AVcAXWUMAB1hoC+cuugCvgCrgCrsDdpMBTD3ba//Rf7bY/+M6UzSqj6JGDGfvznx4I7pqlXPc7arlpgZ0euXsACTiBuOkHApCJxQAi7RAMOHNJ4eJL8ArXCtuwBWHqFb2P/aw0wraCA4Csrqy6EV6et927u+WYISD82jsAQ4SZA3YATe0CNbz3Vg5gBsCJcPhPP9pj33xeIeKCJ4R9U6q4d3vaHjvSHoDPgqAaWWA4d7Cb9csxtmc4HeYIzChorlcFNy6PVeRWi7KjOnLqiqjXgD7tAoO41shxYp0dAlnxACK9qYyt7744K3gT5Y2dONcaSigP7c4GoMfau3ReGJybHsEztGEusatrWGVxX322PwCyF9+alyOuolJIgSudiy453e7bqw6BWnOXSjcJYB/sp0QuKiXcJeCTVTke8G2XAA6unzkBM3xoaXWBBF7dL9fg7XIAHdkjaJQbDNAIFxYdDXnuRuc8wFBda1w7uKLoJJiQu3FEAfxcMwy0Btg1D8o+KVXcJbffb/75AfuPz6F7w4C1h/e0K+w+F9YcTnXzG/UzLjw+F1md19s5fnps3r778mxw23GcYr1mLxybsyE57X7ls32389C+b1fAFXAFXAFX4H0FHGC9L4X/4Aq4Aq6AK+AK3BsK7NbN+N//ayMfmcXiiIk7+q00aTKpavW0XC1lfUXuH+BHRiV/AI7IqXTjG3z236oWh2U5iyhVXD4AEMCW2z04BjDqsODOQE+bvX22HAAcXffu20MHQhxTyj5Sx8GCnDuEmScEdAgzJ5ML6w5A6ZIcQO+cLiqMfF5wpCbgJ6eZXGTbtqTsyN4OubkWFAxfk/MryurqyF5bGRlQP5Ebii6CMTSpaN+va18HdgosCRoCyXiNjKcOgRZGBPyu17lX5+bZJ7oFZ9Ih3HymoNwwvb9Xri8C54cUSB9DoR1yLh3VuuNBphR5ZjuHBLzau+z8lVpwoqHL/douLp2Mt+eREr1Jhahv6U2FskZHayMAAEAASURBVL3m1zbyM9fNdmnF13oH4A7HGI0BAKu9+h3XWwz0VtsP1xY6ASMHe7vtsaNdoSMmJZ/ALYAXr3FeCb9vHik9x7V+Owfuw2PvETofOTRDl0dNY0Hg7PnX8/bxB3NLJbfXn/vbOSfftyvgCrgCrsC9qcDt/0vs3tTVV+0KuAKugCvgCrgCH5ICAAIylyjforSOQZlXXOKXvIE7JSpZu/U33sAdgBIghPLA9Q7mDxAqllut0Z2yTzycUslgm8LKr5VBAjy6BYD4GlH5GCWQOMRk9gkaUJIG2Dp5sSQAJOijw1NqCRAZV67TwJTgzvYIzMzN4+K6PkOpKPcPzi/KFhU29f4oK7NqRtBr38724DJ6/4Ub/AAY3CK4Rn7W5FxSgHExZI4N6rmEngOOMDhXMQiLd4ejibXW6kDFhB0UxIthV7xN/Ej217d+OqcSt9ngxANuPXSww37ji4NLYC3ecmOP16AgYexrv5drsSO7jg1X2FXsJuM6RrPmwWu7FfZ+6kI5uK54rVV2xD3bm8hj8xtu5c8svCU6T5w/nHY84v0DqB07W7QdKm1lfitBxVs5Fd+XK+AKuAKuwL2tgAOse/v8++pdAVfAFXAFXIG7RgGcS4ASbq7jgTtlpZvqyN2jDofaFDcRWVC3agA98sqZWmDnGuwfAEP5440GECqUKqoUEhjVK3BVlBtqelZOKEED3FLADZaXrtLhLiqxA/6kxaM6FttCtzqlgal8jdJJdbSTHkvTEHAgOF0ZWfmaQsGTem1BsCXxAXdbSvtKqc4QuEepImV/jBZ06khuqLsi72PelBPytXys1amR9fK11nj7TNm+97PpJReawIqaav7sWD642b70dM91b49D6XGrBe2WAUbOwTmFzx87Pa/1CxxtzyhnKhnOIUDt5zUGVGbZnsnqeog6c/ZT+nqb3VeslTU/fqTD3jlVVvmostd0QXFNVaTTxHTV/ug7k5aSM/BJOcd+5dk+na+fl0J+XFfAFXAFXIG7XQEHWHf7Gfb1uQKugCvgCrgC94gClLD16Sa/LEcSN9g4r1aCVziW6KgWD0qiJuRMEvqymfyCdSl0nEyotYBT/P7mR0q9CC3n+PFg/xyP0sTQ/U9QSf99YACmgG8VhX/HoyYINZ1vCGQthFIyzDAVxUTV9Tyh8oCxIWVHDSiontfIopotRCV6rJ+wdBwz5ICF/WvXi7JWkesEJOP9dAFsHj3KlzqwK2NvqWwsJTgG0GGLg3Je7VWw/Volcc37+rB+futUKYSmozXaowXjleMFe0a5bwBKgF5e54Gg/0ajYRev1kL22RaVMeLgY9Slz7sq2/yzn0zKyRWdx7dOF+wJwRmyqMjq+nkOOmG2b/lwCRGfg8fv61LmWl2QUKWlRdFBjaCxXpvJ161Vjr0fvDyt7LlW+/xT1wPDn6defmxXwBVwBVyBu0uBn+//Ct9dWvpqXAFXwBVwBVwBV+AOUGC1LnHx1ABczQNAg2OqTY4kIE9ckkdXv43CGgBIM7ziOHScI4D8tXcL9pZyqciiGlLJ1Wcf6wrAjdI+jhtBrethEuCJ8PjLV9VtUUH15FalUwk7vCtl/T0pwaqGgF0jOM/IUcKttE1OJ4LOJ7Xt6yeKIXtKBXCWlpMmJcdOKqmMJlGJAzuyAjdJBb4Lji3lWDFfwMSzj3cLbqXs9IWCfFstyqFK2jMP92xYD/b3YYzGgjovLsErjofLivOA1nT9m5SLjTyqOcFAwsd/plDyhmxt7eqEuF/B9H/uY73Wm4u6Tb74dj6E4MfzBiq+eqJg+7bLhaWyzZWgaLztnfiIDoBTxmYdZADAX312wD7xULe9e65kb6g7JXAwunZ1PS22yvXXsJfenrdPP96l46xhN7wThfI5uQKugCvgCtzxCjjAuuNPkU/QFXAFXAFXwBW4uxQAEAFsuJleT3lYvPr4RhzAspH3xe/nMYJJhIPT0a4luJAI86G8rnlQPgc8AnpsZMTOH94DRAE+4fx58c2Cnb5YDOV4bINrZWqmZp/7WI/yra6BskVBlVZZyegux6CE7/iZkn3nxRkrq5xwUe+lJHF2LmNfeDohZxclei1yadVDEDjvITtqoDdhn36sR/lZCfvxq3kFi9fVOTFh2/qBXCoPXFr79Fw9AJmpOXUuzFfk7GrY0NbFUFb49APt6noYZSwBxJrXxnHWO8ana+pAWLNurZMOfQS/t2oNUQnkevey8naRvou2c0vWnmvkBQEXQ9dGrhXmu3NrOkAWXG0LCy12/FzRnldAPWWeOIsKizU7fmrBhtWR8fGjqeA2I+dr+WhIl6vTC7Z3x+rXQ3DX6eWNQs/lx7qVv3MNX1ag/8/embPzoxWd16R95rFOu08dHDc6uM52bE2Fa+v1kwWbU16bZAlZZrVkw9pNJa8h02x1jTZ6TN/eFXAFXAFXwBVoVsABVrMa/rMr4Aq4Aq6AK+AKrKhADGO4SacZGTk3XXKjkCG0XrABVADcxG4QDgRsocvaWoPjR6V5EWgiV4njr/fY8f45flFQ6cpEVTfgwCJ1d0u2hRK8ge54q+hxiSFd/+QavwHWmBuuH4LQAXXTypw6f6UcShuFVaxNf321KPz6ykTFLoyWrTPbLrAQ3fQnRFWG+lrlCBMZEFSbEaH69k9nQme7sHJ9A64R0P7ARIdt35qUY0w5VUsOm3h6uNBG1EGvK9ctwNAqR1Fd+1gI7jK2qcoqdkVB7wMKip+ardq0nFxtKqFcUMZRo3XejuxuF/BSJ8dl5YXx/tfzyDy/9cKMnDrzKjtTjpe12i7N97GjnSGknHNPCeNmXEFVQdC8ziPrrup6pHTygf0dcriVJRvdFRO2R+WOTz/UGaAl82XbE+eqQQOuQcVbGcuraN0Xxqr28MFEuJ5D1liLQu+5YJdGq85pd4e6Ly4zFrHJ6GTV3jyhNSpfLJtq03FTdmhXdsPXZnysW/WIc2xsomZff27KTpwvhXOuOH9779y8/eoXB+yTD3Vt6lDzKsOcmqnbrK5RVWIG52BGwfWE8uek0WbO56Ym4m9yBVwBV8AVuOcUcIB1z51yX7Ar4Aq4Aq6AK7AxBSgTogSuUFyws5fLAj+CHW2LoYRthxwuQ3Kv4BcqqhQNdw0wZiUXCjlQzfCKWQBVgGAr3fSSJzU+VdeNtzKtBD8yTRngOGhwH3UKYq13AMEAWBXtb1bw6j88N21nLpQC2CAA/m/+cr/t6L+2t5XmdO3V1X8CrIUyRdG1RYESguUZVf28KCqWVLlVXU6hhCggoOsaJon2mZbTJTeYDKV/py5Vtf7lW6gkTNDmjKAYAAuo0t258p90rAGNAEjAtZrmAliCOuDuujBWtskZgatFlSIGV1rDcrmaHVuYlyOp86YgzGsqMfum4BuOpro0bxNIm5yrhrI9rhEldGn/jQ3lSgFMLo1X5SYq20tv5i2ha+fxwznrzLXKLZaTsyhjc7pOOwRFt/WnFHp+LcsMIFnXt+hsRFqzP8mi9asMTvtqk8Xt/gPt9sIbEliQDIglCeXkytgBlRouH2Sn/eztgropAhzN8gKDJ87h7jM7tML2YaMP6VtN1/sZhdGfvlRZglfRgWta9PdemrWHD3Ssu5tkPGXKbb/9wqy9JyBGUD4OLEZN11SvNL9vzwc1irbw766AK+AKuAKuwM0rsPJfOze/X9+DK+AKuAKugCvgCtwlChBmTanZScGeSypHAgThwCrKccKdOuVhk7OCFKrPA8qMDKZt/450gFIAq0m5NQA18wrPBnDhuGqGQ9wUN/+ObFPKLPreSzMBnIEc2gUX7tcN91YFlseD0rxO+YbWO8RQ5B5T+aCcI//hh9N2XF3m4syqy5Wq/Y//9xX7238+Z8PDkTOMkqnNDFxhlMrBrUpa9/is9qKf6ST4/qAWUHCkV+Hry8shyViKoBeQB5cYpYTRO5tRFoHeQKFd2zI22LPyn3Q4sQYEGCn3atFckqJdCwJo3XIsVbEuqXOhPHWhXBIoCPQoCcZcurqoLK2q3GCaHwRnnQNIVNT7cZB984WCjcrpRhA91wHwLldvtXcEPx463B6yuwii57XlGqx2uMtyrX1D4PE7L6ssslQNOVfPvTZnv/aF/pDpJW+g3HStIQyfsse4y2EMAXeqBO7ylVIodeM8EWSfUF4TjilywGRGs08/0mU5ZYW9c0aEVOdp62DCnrq/K+SKNc+L6/bqtDo+LsGr+DWuy7Gpmh3YmVkR5Mbb3fZHrW9uXh0nY8rUdMCK5kw3Q67T9QxdqgHKnrxQtB+/PqfPvLRveqP4p9X0bUvfytdh06b+oyvgCrgCroArsGkF/H9lNi2dv9EVcAVcAVfAFbg3FJiXswogMaGbdULEgT4yEIWsoaocVNzI9gugAAQqgiLvnI1yn/qUeTSuG3ncLUAKSvfYtreLrm/JAK3iPCxeB7YAjfj5+z+bVYmSAIvGgsDYhIDWsZNyGnV0WlauGgbH28hIaNIqeBRMW5CDpBhKIeObcB1Sjq5F+5MfF+3goardd5PuGdZBudpbp4oBImBRIywdoALkWdTkd23JWFb2H9xaSYigBhrEMIfHXnUE7GxPBOcY+2DJrLtDpVq//KlBO7g3HTK0wptX+bZ/e0bna9EuqkxuURr0yq2F4+z42ZK1aJ6LM5TXRW8G9AAb2zNyE8m1ldTrbL8cMK50KNZFOR3nGdA5NlEUJEPhSGUcdeWlMkYcZLbkqNvIeTw/WrUfvKIMq6pyqqQDUHC+WLP/KKj1X/x6VmCO0tKkrpEWlae2hUwrYCDrnRHEu29vu5Xk0DqlQPwyAE+vPXakyz7/sW4BsVJYFvN5/GjOHjrYEa5doB8Qcfngur4OSsYbaLmLciRFq46f/PAfAXjDCvRX00l9iKLjM2egHDlo82W5G2vJ8Jm70ewAdeek+/h0VXlqs3K/4QqMoCrXI+vE1Dcnd+Mb75bswQO5cB3faJ/+mivgCrgCroArsBkFHGBtRjV/jyvgCrgCroArcA8pwA0q8OWKnFY4S2bJkVKgOMylpbUtdKjbWU5ZQfFDpVLN2mSSmtB2QAP11wslhdw3kzeV0Pa4kiYFTShtw83VoxttQBZfnR2CNYJkBUGQeBBa3iqqQMc94FMMsAAUGxkpwaEuZS4tjivnSTflrGv5GJ9r2BsqfRtWCHq34BFAaSV4sfx9y3+Hr5y8oNItuV/YR293QqV5rJfSPQEiuX16BfimBAY7OxetQ8AF+MK28eDnkaG0PXooay+/My99RSE06S5p9A/++jZ7+uH1BXEDZPbvzKrcMB1C5TkXwLFtA5wzuvNRyiifG6BG27boDYN96TANygx5rVtzX2sAr3BcsX/OX4/WfGGsFkE3vblFO8d4tl1QpaGNkjrvgL64xHKt/fP6yQCeopB15tkiIMfxAKx5XZdb+rJhnzkF1rPvbLSMsGtKO0e2pK3v44ngGCyVFuXWUrC9nmNfy0cE7T74fLwdYKtbpYs81ugOsDRYT4+unY2sK37vrXxkSQ8Lwr28L2evvzsXdg1kzLUn7YhK/brkvqI0uEefiZVcdmRo8Zl842TRLo1V9NmvqgRUWWZQO2m+ZCIM+9Wv4fezl0v2E4Xkf/LhTkHPa9fyrVyX78sVcAVcAVfg3lXAAda9e+595a6AK+AKuAKuwLoUAFTR7a2sTnazcrFMqqsc3ca4GS7JYXP8rCk/SC4XOauyglet+utiS09dYCAVcp+qMrYMD9H9rkUASeV5ciEBHQiD7xIYaWYHHAd3Upu+MOnEI63Ab+6bKfcCZmX0e1weFm+znkecYiMDbZbRHKrKwVo+lP0dSh7PqmPbsDgJc+4QbALgrJTrtfz98e903iNXinJJdGI/hF7PKNB9SGVWABNK9aZUXkj4dV8XYfiKURc0iMEHbqhDuzMCXj22fTitMPiG9QnCfOWzg3KwbfxPOIBYMyCjXLBNXOOB/a125hJuubKlBGP2jGRDKDrzAB7iiFtrkG9GWRnlggwcVoMqj9w6UJV7TgBSJxwUNCJ4Fcr15CrrF+BC240MQGC4YLQ/TU/XmoCgSArXX5+gI4/kXgGvlo/4XPL8yGB0XCDVZq4j9gHYHOxN2cFd5F6VA8Tiut2mEtoje++MLCjW/Le/OmQ/ejUbsrpQ5YDKJe/bK1eeBAQg4wDMLTkAWRe/cz4n1CXz9MWyjelapjy4FKDyoq7VRABfDZVbxhQ4pdPSLbCKw+vspdL755n9+XAFXAFXwBVwBW6VAhv/6+dWHdn34wq4Aq6AK+AKuAIfCQVwUtC1jhv9qm5aucFV/JAer5VQ4Wa5WKqE9bTr3r2iH1PJRZUpAb9UqrQg54tufAMQqaeCIwkAEcOaWAgCxnsEi/oEmkYnrrmw2HZYDqJ9lMN9kE3Eb1/zEWAxooD0Lz3VY7//nalQ+hS/iZvwTpXO4c0q68YeWAcQKQjAUWqFY4oOdesZ4kDvD0AHCGhBQIp8JEZNUIiueMwHp9bs/IKcbFXp06rw8aQNC/5RukXO1cOHO0IgOCYfOAPzWAmmAb+AZeyX11eCOOHgS98Gl/KtcC/dfyBphUKrTQtW0NEPSBh3hwSC3GgAuAjjb55TX7fcZYU2zbvddikrraHsMTKm7j/UGTocDgo2NW+/2v7J5dKygk5cK59RPtU3npux0XFdWFKV64Lw9fv25gSSssH1c6Prg86H6EIZLO+NXFarHX3t5wGph/dojcMKj9c57ND1g3NvswOX4yty2/GZ6dHnhSD9Xp2Pmx2flG64sSgP5RrJC1BRQkmDhEotFa5zPnuzBQX7TxZDueCM1lMTZKVpw7TCwfK4LiUu0HGoPx1cbFzPXB45AcnhfpWa6vPB1TIjp1ZFnyEaEvhwBVwBV8AVcAVulQIOsG6Vkr4fV8AVcAVcAVfgblVAACGbalMJnErqkmW5oOQqUunRivk/0qAo4DOfqalbHFk7ct6IFABJWnWny40uLiecVPrvAwNIAWR4/L4u3cjPhdwtwMzwQFplSbmbglfxwTLphP2yXEzTKlV88VhJsKqukHhyvJICRguhzJCpEfAdD6BdvqgyvxzljLhvbnxjvk2QDCcQwIBBxlSSkkBlfwGhJhRsf15dBHFBMZ9nn+jWtq0CCXQG1I2/YCHgCIiFJjFMYl/LwRTbUApGVllRIHFBJwanFfBtqzrxAQ1XG72CJHyRT1URaDrdUlCJYls4tp4KLjHKy0qhdFMAU/PFOQa0qghglvSeSXXii+KklHWmuZCbBRg6INjYaMR6mQ2pZJGwedayln5AuMsqRyRDi/kDT5kXX//1bw7bv/n/xu3ilaquKVxq7fbbf3FoA3CRssXVFNn48/H5aT5HG90L1xeNBf74u1MBLPUIUh5VDtvFKxX79S8OhpLTje5z+faU0HIcrhXyzS6qLPDld4q6FgWn2wUXd+MCbFha0JSsMgD1lByDF5R5Na0mDaBXzkOfXHCH1BggX8rY1amKwXQ7dF669VoECPWoklg6T6aTTR+i5RPy310BV8AVcAVcgQ0q4ABrg4L55q6AK+AKuAKuwL2mAA6iHgV5Dyj8eodcUNPKGgJu3GjgKMplBVMEMzo7BLKAVsorIvsI2NGjG152QclgPHDO5HQjzRjsabPPPq5gbd1I4wDbbJlXvO/mRwDQ8JakPfVQp3XrZvyq8rpwkqQFStKJqm0fyoTg8mbnEcBmUiVVEVCKwtbJ8FptAGg++Wi3vXwsH/RKat/bBXQAMq+8N28vHZsXAKorCB8nUDVAp1/7wqCgk2CXANSkoMKu4ZRgAxlP18Rm7s3zwjWDQ4Z5UfY1J1AYYJRgRVSqyXqagqBWmXCcAdWtc0V2GQ40oCMAAihVkGUJRxowiemQh3T6opxjKmvsVA4UMBPoQh6UthBMatN5b7V9I+n3M8tWOfQHnp4WNHnrZFnnXg44rRfoWanpetGBk8lkcOH9k9/aHhw+gLJmPT6wszv8CdZEl86vK4T+a9+aEMSMCmfJgZtRl8CnHszZc6/n7WP35wK4BBBtdqAlDkKy5Ghi8PXnZtVFsRI+h5z/k+fbbP9wq23flrC6ymeBoeevli2v7eWDC+eY5wCzD+zL2t7tacspMP/512ZUGtvQvgWysm3hqz2jDDBdKz5cAVfAFXAFXIFbqYADrFuppu/LFXAFXAFXwBW4CxXAYbJV7hkymSrVjM3M19RtTjfagjqrDUoNa5QdKkybrnCUpbWn5czRPS2QBrjCAEAAjxTHHRxDzTCCm+31luytNo/Vnqec7eMPdSv0O6OMn6rcRwuWahV0yxTs6IH2UGJFKWEYeqCUjbK62HiFk4XubjndsK82ugW4nn2yJ+RaAZdwL11Ud77f/86ktIlAhZZvLfprDLfRy2/P2ee0PTChoTovYBW64X6ijI7flwMMYCC5QwCskuYUwy6cXVmVs/FenDQZubHWOyjJ49wtL/8CjrVrvWOa63NvzNn5S5WQjUQw/ohynwCQI4MCGAJOzLtPIGz5fG80B2DlcXVGfP3EvPK4KkFv8qT6e6NSQa6FABC1FgBhVtfTR31QdkhjhO+/MheAnTL/g45cZ9MChqcv1QQga+H6ozEA12G39N7s4POGau+drYSOolwvlBRyDdFR9MyiHFa9+l31qsVq1E2Uz0GbKDbXX06wkhJWuohSOkk5YWtLj+DXvLZSx0q5+eg+ivuKc6U9XZe7ttl5+/tcAVfAFXAFXAEU2Pz/Arp+roAr4Aq4Aq6AK3BPKACEqNTkppGDiBIuHErjU6P2zpko82olEXBtVReiQG3cVu3aBzfPvV1JdX5LBlcPpWk8l2wKkF5pX7frOUoGyZzauVVEDVil0PVioRJKpOoiRpT/EbCOkyQ4tAROcDzRlY8Mq2RRZX6DCpQX8LnRAAACHYBY6MdtfUolmTiV+C90sBNFGJ+WI0sAIyVNCDhnAPRu5PQKG+kb8IiOgcsHkI1yPG0RXhKnCPMPXR8Fudg/7jaynJpHU1O94LiqhawkOcPkjvrhi9P2/BvzoRTNxO+2SMeS8pFwXFEmynyHVLrYDCOb973Sz8zr3GjZTl9QSadKFJku5+CKnF5cJ2SBdXYAsqJ1rLSPj9pzcTnmnMr5inIiRi44CaH/KNeDWs4WqgEGxmvjPZyLzXb4i/LRTLri/lsI1xsAK1YVCEpgO7CwVXQU3EXHTJx1XLuUh/LvAXMFIAJwd8jNONTfHUpY2Q+fFWGx4AzEuci2dHuM3hevxB9dAVfAFXAFXIGNK+AAa+Oa+TtcAVfAFXAFXIF7SgEAAjevgJutChjHZUFp07zK287LUbTSoJwI0097Rt0HddMbQNG2jO1QNlS4kb2el6y0iw/lOZxCuLHECgSnlPNTjToc7lSpJOsNoEc39QC5rCDKlMonq3KmcFOOS4YyO27acS2tNchIClrKoZKfx2Wld8ArRJlUXSnIlQzOopAVpYyh9QyC54E6uGOADnXCtjQADQzmzARxYgHicDnh4qHkq6x1AJmYF4Aoo9D9eLA2Buu7NF4NgIqOdBfGSvaTN/KCcde2rZRrAXLtGkmpE12bgEvUcTLaw/q+zwsWFnSsmk4EWgP9gC3Mt0DwuA6XDo68pYmtb7d39FbAUNbZqvOflVMxIViFywplWSVfQ/1ZO7Dj+hJQrpv1XR0fXD7ONUoAKdXl/xYFbcPxdDBdRuEawnk3qE6ZXXLS1evl4AZkVlwrOLWAVtsGE9ap1xlAS+ZD8wVy0QBgDcHrS/q3ARhZkRPr0tVWubZStmOITLa1Pythx/7NFXAFXAFXwBVYpoADrGWC+K+ugCvgCrgCrsCdoMD3fjpr33hpRjeVi/alj/fZMw/kNu26WM96cGEAbArKuOFxuVsCOMIXg9yq3s607d8leKUQ7fOjuutuGmndzVIm2KdOc7u3pQVHuMmNnEWUDuLQ4cb9dg+6oFEitZZDCqjEoOyxeeBM4gt30KRCrGdU0kU2Eb8z2DdrAbKsB2DxHuby5z7WY/9WIeR0dmTgQuvpSuk899rRPdn3yyvDi2t8A0h0CRzMzSs3TF0Fxyb1Bs2vU4ACVxVQAncPI+RkKcD79749ba8en9NrUVfAzz/RGbKqmh1cuG5m5LY6JWhFOeMlBaaPz1RU4lfU+V8SIOxVuqmMcXxa8EmXAQHfNyqrXHrLBx6AcIuAOGCI9MDxUwJcCfKUA3xTF0p1ZrybRuxQAyY9/UC7/ccfK5ReIepAID4f27ak7Rc+3qVr7HpcFb9vM1rQEIEss0cOt9vJiyUBs+jaABqi+9E9Kds5nA0dBQFTh/ZmQx7ZxVEy28y6BCd3Ce5Gc+I6wB3XGkpumQ/wijGnY7COvK43Sk8v6Bo5e6lqbwsY37+/Xa7H66FceJN/cwVcAVfAFXAF1lDAAdYaAvnLroAr4Aq4Aq7Ah63AP/3Xl+x7L80JCERg6OV3CvblZwbst35ty5owZjNzBV4RzF6RW4ISJQAVziPcSVFZ0/V7pXsf2Te5DrKPUiFDZ4qSJAGdNv1lkdTNN+Hf3QIrvSoro/semVHsMyv3R6/2ezvHvIDHz47N6UZ9Qa4RgQ/lKB3VjTgOns0MIBAOtNHJCF5RVhU7norq/Lfe/UaZQIv25U/2WUo3/YRoz+sGf1Dlds8+0RU69BEUn9ZrHdIJJxNgYa0BeKDkEHYwJOcW4A7wA2hD93hwbv/7/+OSnblUijpIal1jk1VlWRXtN39pi+3cIsfW0qkBoACm6MQIZwsllNIPwBkDvHi/4D9gHl0HKcvczAAUdut6mpRjjwB5IFl7Fp3b5EBqD4CNnLKErru7ZQBOAcVdWvcjh7vsqsLcX31nXo0EFgTr0vbFj/XaTpXrNl8DbL/ZMkrOPzCzQ8CM7pSfe6LHfqbj0Ukyq+6XTxxptz3bdOyhbAC3XOOUKi7KTdXbrc+1Pgi8FwfXfEldL8tJwcrIsQgppXNhMhF1ouTzPzpRUvfCRoC/dD6kgUNW/xa8ebKk85m1r3ymJ3yO7pbz6etwBVwBV8AVuP0KOMC6/Rr7EVwBV8AVcAVcgXUrcEIOlx++nA/wKkYPNUGIb700bZ95MmcP7M8FV826d7iODXEQATyaB9k1OCgyKwRlzxWU/6S/IHYNpwNo4aZ0rLNmJZWVVfU+yow6BRoyAh4TAmPDKjsEeHUIUuAY4kY6vYFQ8eZ5rfUz835OgdizCpqPx8kLxVCO9fChjvipDT8CbQZUOtkqRxwAAajFAErFzrTomQ9+Z04TAlOAQdwzBKE/+1i3fU5dFucV4s0+zl0phxt9QBAaUU7Itrir1juiyiztX+9jzAuuxYNjvPz2vJ29XAnAkfXQOXBR5/2UcqcmBLLadb76c9euA/YSnHh6DCWJJbnodN44DpcL+2DwO/N85Mjm9WU/u+TWA8I1LgmS6TGVSgmQJgNI+/7P8gEifuKRbpW/ResDvMbuOd7/URic32YozLWTaFuws8r/qunzdmhn1hIymnHWL0/UbP/OTFgjXSwBQMDJUqUWrsE2AbCGzisXN7B5LVcjTigaMeRV+pvXdTekz+XnP9YVct4oC0611QSYqnZoV0af+zZ9ThXkLhg8oI6gHdlsuG44li7PcG1Ny5VYqSrTTRdDSv8GdAlW8zU+06JrWV0k9d6CuioSBl/S2q4qz2yUskKtoU0foK99Y9L+2e+M2O7t2Y/CqfM5ugKugCvgCtwBCjjAugNOgk/BFXAFXAFXwBWIFXj+WEE3mdWQSxM/ByeYL9Xs4ljVDuyMyvvi127FI6HQK42laqAPvBRYlybV361gbYGp4f6EXZE7KV/UDauCyvP6AogRCH11siKHkoXg9+Dc0GtkTq235O4DB1/jCUrZLlwt2/Rcw/p0U98psAJouawcpyO72+VuiuDHGrv5wMtAJWAJuT/cmAuxhG1YR3vmxvs8JWj03pliyJMiVJ2Mq/vVwY2Svy7N8d1zZYGlsoK1BQFVNkdu1TY52yjj2wjAWj7pZucO7pjX350TbLj+XPMbcIusK0gIHQtjyNJOZtdcq7XpdcoCCzq/IwPK71J+FiAFCVpEWmSaUrfFXhuRa+hmBtlLR/YoJ02lgrNy7BQE/H70yqydkD6lSj3M68W3C3bfvqxd0fmc15q6VV73WbnXgC4xGAruM01kJaBT11rI20IGnGtck8063cz8b/RejndmtCT3mwCOrp8+AaP9O9OaY2vI+qLkrkNgrt7Ql8ARgHRGmVgXxyoho4zsuXJwv0XX3dRc1S6orBOwCOjE8bRvRya4HleaB5lns4VaAKnjAkkzhah0kM9qh/K3tshpheNqi7KvcEky+H1iRg4qwVegE50sAZqMslxbobtlOpoP7krWCEgjsH1imiYFKgvVcxhJS5QRjtaD24v3y8toF8eL9vf+2Rn7g39+VGvgWR+ugCvgCrgCrsCNFXCAdWN9/FVXwBVwBVwBV+BDVWBLb0oOB5WDmeq3NIAv3CIm5MnIqHPd9FzdMgIgsQMobLSObz99K6+yukJ4330qp3vyvq4ASHhrQjevtaU8puZdRY6ea88AuoBUDYLCNYGGpkjpFx31OtVl792zFRubIEhKd626AZ/UjfKi1sONcL+6D3YrvBzwAzC5lYMbZ+ZW13z+569dtldUctlYaIQ8pV3K8yFwPpvWRtG996YODewAFHADz3oAIUCt1cos44MA9F56M69SvYrmRLYR2VLAmFZ7TDANlxTh6GcEsKZmdYMvt4rCoFSuWLUW7X/nUHrD5zo+NnlbOGgAOkDH4S2Uo80F0BAuKp0j0Bvnf9d2naeliCl5hMLzgypLLGr+09poaCCp805JYZdcNlVBt/ngpGvPJO3pBzvst//ilk3PM55v/EjpYLHSoq6EAn8XyoInKj/TAtD7vCDQ8bMlG+qLOuKRSfb736rZr36+zw4LUAJkgCkMABaln7FLi3NGjhmv87kCXLE9pZpr5aTFc9vs46lLFeWOcV1G+ValcWVDCdAdVNlli66r2AFJDllcusmJAtIBt/i881ljMP9RubOquualSrguK8qyOqdr6Oje9vfXG88VEDgjeDWvtc4I7LJurgdGVp9dyv4Ahlt7rs2D166qLHhW4ElT0nUUZXMtLBAAH5W2ArSaR/gMNqL5ULYLNC5qrpQnz2oOFV3b8XF5J3CLf8/+1b+/Yn/317fesuuneU7+syvgCrgCrsDdpYADrLvrfPpqXAFXwBVwBT7iCjx+JGcj6tR17jKQI4JXLGnvjqzyhRJ6Tm4Z3YhyY77e8bVvTQQny4JuiLkRPnW+pDKfuv2FTyuLSSVMhDDHN/3xPsm+IQQ8HsxlVq4NHsm+GpQ9C0dOQpSL8qFGvUU32gWxFx1Db9I9ubUrLwdYtVWd1Hrl1mIkBBWAQbdqUP5IaR7z+to3J+yt9yJ4RQe7su6WT14uWa6zxZ59vFculZs7Lple5GuhFboBs2JHymrrOSsIQ84UZYQMQq5n8wv2zumCnGsElZtdlGMMWFCUyy7aShsWBRAEA0cFAbZtMryccw1gA5Jl5JQB8GwJQe/qtqjXgAmE6x/dmxZgTGktghXijzpFYdC1cM+IXiskrNbI2KNHWuSmWQydGKflyplWxlhOUGXbloQAyYIceVGJaPTuzX8HVOGMAm7UBGYAlPGcyFIqV3D/CShCYvRCQe6s196VngK7zXA0BJYLwJAPxsgX6/bdl/L29Z9M2YxgIU7AX/xUj338wS65nxLKbntf/c1PfoV38rl59Xg+uJ+QNqO1ZQUvCTgfVxkegeZ0b2R+FV3IC7pG6NSXlC2pZwnAUf4HMEIbQJGqCD8wAFpFfR66luxMXKc4p64Au6TjjPYxLSfWjI4pfhqAJF0sW7SvXFbgTM4vHIw7t0sH/QdoRfctdMTU73OaX6m8KOdgqzoOplYsL+ZchfpHQWzeRwkywf44ND+grvaND+zslUqY47DKRX24Aq6AK+AKuAI3UsAB1o3U8ddcAVfAFXAFXIEPWQECzv/x39pu/+vXRu1dZRMtqKRo/66s/bVfGlA5mXJqBE640Vy6S1xzdmMqS3tBDiDgFQPQU9aN7qvH5+1j93eGm2c4AMedUz0YkCen0rHl7gpcFJQw1gRiUgJQlLnxhYdnQMHd4wIaWZXSZXRHHHc2S2iurdTMCWqVVP6EA4tSNNZwKwZQCDjDmnAaEYBdln2Fm3tAAV8ce3J6UaVp7UG3mzk28KBT2vC13lEU5AMCxIOyTMAAYOHtM+XgrqG8qyR4RZg2rpTQsVFuKELxpwRxNguwOGbsQmLOOHB+66tb7Y9/MC1AWhLMbLODurb++lf6w3lJE8DdPFm9n7K8bjmUmgfXyjmdws5cI+wfaDQlKII+MSxq3n4zPwM+hhU0Tolmi8ASziSuvQCzgC7NO5VORYE39Fteioazic8L5/2Hykb73T8ZEygkUNykrdm//uOrwYX01IOdVtHn4rr9Nh9jkz8DpX7vO9P2wuszATztkKOOElIliUlz7VRuO5x9R/Z0hJK8hggiUJYujEd2Z22LSgf5fKYElLneU9IYuBWIkr6zXpxlZFuxRgAjY1bX1JzO95Q+lwCiS/oid25eOXW4rQjnx3G3sICjkGw3ud4EvNqkNZ8p5hSXyXIutg8ldZxkKFk8oJwuton+HQqHC9/YRwynOU+ss3WajqRtalzQE0plS3ofA50BrGn9ezHUq6y82UrI5GKtPlwBV8AVcAVcgdUUuP4vktW28uddAVfAFXAFXAFX4ENRgHInwtH/29/aKTdF3fIKTCe3iRvK+OZwIxO5pNBkMnWuG7rhzwNMBH3iAXwIQesq31ruKsJ1dF43wJOz1QApKLfCeTWkvJxWPeIM61Z4M7k+2CyKNZUXUkmoO9ScMoqG+tKW1X5xRUWZUTd3l4qzJgRE62aYbCdusEfHyfepRs4UzQFQlNB0uKkf0c03a+K5D3sA7Trbk3K/qIRQdpOS3GIQt/Z0wqama1bTnX5ZQdd5wZc5QQeARNBIWrG2ss4dgGgz5755rcACOgRm5ZgapnOk5gBMpISMUkXO/6JqQlUptuYAlgBNmjOmwCaAuVsFsLjeDwngvPxOWuWUCp6XHmI4wf2Hfsy3lSc06I6H2wewyrpWGoCsP3t+RqCzToHr+24gXEp/+tyMcrWUjyaA17bK+1fa542e4/okMP9//+Mxe/dMIcA3ws4nput2YNeC3HBZZbQlwjkhr80Ejh4+lBOAUpmurhGy0WiOwAAMkeVGCSGdGIHMPYLZBYE4PmNkejHtlC4eygUJTb86WQ/ZVT95c06lvCofJENMtjvOPxlVVTkma/payETOObK06oJf83IEkl0HgJqWS43yxqzAF9ckwIsyZs5NTu5KmjnEzsIwR7nFuM4YAYBpUnQ75PO6Xe6qv/lVs//zD69YlX93tF1oBLEtqXOXDnOPAV20B//uCrgCroAr4Ap8UAEHWB/UxJ9xBVwBV8AVcAV+rgpwc054N1+ppNwiunlvHhtxEfUIKlEKWFhyPsT7Sevusb9nKfQofnKVx1ffzttbp0pydNQCtNg6kLY96hiHA2OroAg3qwlZOu4/0CmXRUFAJhEcGmRFHVC4Np3kZG4JTqQ4x2eVQ635NDfWOJYAEnXttKybdW7Gj50uyeGVMErMUAuHCTf1zOvxpe54zcBlzQPdog0AOgd3pZTbpFKu6YrNyhWT1JxSyjvHadWqdTCSOufc6FcEhsoCEnl1/KvXK7ZzS0VgMGkD2s/NBLqHYwTnHGWP1847rqVMG3OQi0fa8ju5XbjwVhtoH7tzmrfh+fWMcgAkQDnK5JaIxwpvpEveb/7ioP2ZSv4uCVACdz/xaFpB6BWVutGJD6jVajt0LT4tNyGQpVS5nsDxWQrOJOm6/DMQH3KeTnmCKpnOCBbFz2/2EQfV5YmKynbnlG8mFyW66D+uXbKqTl+ohI6LfcqHm1EYu0xYAb51CgABO1knDieub+YPMMY9hYsO6Mi/B8BIHJFAsja5HNtUfYcD8up0VTlhlfCZPKX8sMtj6m4ozkXnT8qPyaECWg72tykUPvpM4hLj2sskBc8EMS8IVivaLHx+8ip/rOrYHe1y3KnkcrA3unXgs0TnQuAq60PjGF6hGyWglD1yTLbl6yuf6LXHBen+5b+7FM7Tjq0Z267yU8YbJ+Z1zQvsqetiTjr4cAVcAVfAFXAFVlLAAdZKqvhzroAr4Aq4Aq7AHaIAWVd0iIudDtwobqSEbURuG9wlP1N5XQhf17padMP75IPtcsus7YS6quyaYyp1q8k+lNaxgVaXlNnETTX75oY6Hk8/kLMOOTwuKbeJ4GlKwA7vSYebcIASgxvaYL8Iv238G3AgBiVo0SrwgBsIuLZfeU1VwZ8JBXuTSU8Z1kOHOu2pB7rCDTZA7cMelFfu29kuqCD3mgBGB6WXAg3jZAIpjyol+JcWYOzrTgb3zGxe3Rxxr5UqIf/pleNJ2zGcCroBIm9mDctdM6jRKufP1FyUI1arVuWcAwouyo0XzD0ryoWLBwceeV4M+AznZK3rklN/ebwisKFgb52jceV+kQV1ZG+HStRWbkxAp8a/+uWhAH9wVzFnrqFjp9S1UUCwTw63+/cqpF9z4vkFhTvFsAZoQtktA8i6TSVto1dLcpotrU3z0SY2qHNEpSvXtBqA3vQoCAhR+nl1qqxjcb1zxQvk6DNASWZCcIhmDQV9RiakN+eUIPm64BGlgkA9XFIVnYcYDuGODPvRfONQ+t0CdzxP+D/gthLcejRaUEmqPhfjcntxzOCY0zwkQXBUAas4V3V1PeAYKc0nlSQzDrefBYDF+9EmJXi1oM84OVZ0mexo+rwznwhAasNlgzV15wCKlPjiHo2gFiDuH/2NbSGgH5BJJ8VTCp8vyU321ul5e74nY3/5SwMCdNf+XVm2a//VFXAFXAFX4B5WwAHWPXzyfemugCvgCrgCd74C3HjjdOCmk95wG82IodzoVz83oDyatL2r7m2MJw7n7JlHZDdZx7gwJtcV95LKleLYQBSABZBgsO+ak4dd4cB46GB7+AK6cYPaPHCT3Mhx07ztaj83l4hxPFwruFWiG/IFe0JOnJJu5Mt6rktljV/+VF9wEy0vi1xt/5t5HshQCQ6XyEkTA4Z4X0np16bzsFPAgYwiSr5wkRUEHbZIw6zcWKkU3R0pyRIwEBgCe5RV9nXibMFePZGwZxQ0Xpa7iLyieAA1KTMEDFECBoC50Wh2zfAezufxM6UQHs77KF8jYH5SuUlccwCPlQb7wYkzqUYAZKMBsrLKMqIUketiNchGHhvbn79ctTdPzYf3UfZ3erRsDx7IKZMtd11ZYvOxmas8PktPtQhM6joz2YKaBnAI4Mv5QD/mGQ+ulb/z1S32j+SImlKpaXhFT3YrjPwXP9EXSt060jVTlexNDz4flIt25civkjtK5JLrlvnT8ZHyP8r1Xg8NByK3IOWE+5UtxWtRBp2aKCx9vNgf55p/C3A88QUYmppdsBeOzdiYOhryueoT9MKpx78TbMsi0QRARbkgDRTQBdgnVhngFZ8fcq8CGNYxAEmnR+ftzZOcHwsZeU/d36HPevI6PdcjUuSwu/6a5Jx0qgnEwV1tRoOD0cm8zUr0BUEuzlthvmBf+5bZb//a0HoO4du4Aq6AK+AK3GMKXPsr6B5buC/XFXAFXAFXwBX4KCnQfDO+0XlzQ/znnuq2zz/ZHd3YbmAHLQpB71B+Va12zXEDF6AT4Y3KATkmN95x0DMd7br0HDfUNzPCjXnTDvid7oBPP9BprxwvqPxNrh65Ulq7AEYZOyr32e0c3OST/QQQYszrC1cMAfdFgSpMSqPqAseyIweVQKQE5H0VlettH8oIWNWtIMcLzjghrAg8aD8EugPj3jtXtUcO6X1N2gExZuXgiY9bFpAAjgEkVhvMh/LAirYFJpK3BVDSka67Lij94hzfaHQLtJDHtLAIAEmE8r2kzgXgEvgFcGoewAm6RdKR7+0zRQGTqLsgz7Gst04UFOadVF5ZWuBmjYM373iFn1nbSmP3SNb+5T/Yp+D2cbswVgwQ7pc/2yeHXIfAkc6XXHy3YpATBUijq97W/oTKCZcAls5vTtlnR+UYm51XJ0hdM8GRKHiDY2v0ai24FmMAyLnlXMSfISCTjG/aX1VlldP2TWV60ZFRZEx5Wm32xH0527cd/WjC0GbbdHwgKJ/hfoL4db7TOul7tymEXfrj8gqOKsmFZGOa5/GzVXtXXUoBcIxxQcfzV8ph253K5uMavpkBaMM5Kc9eKKcEPje0ftbFevkYXbhcVFfCilGq7MMVcAVcAVfAFWhWwAFWsxr+syvgCrgCroArcJcqAPwAWGx0DKkM8JycEv09LQFYAGQoJbx/743BEFACsNRQlzNuTJeDp43OI94euFGpRk6U+Dluio/uzQbQdvqiwunl/BoaSNgjBzriTW7bIwAmhkjxQebVcbFYVokgd+MagB1KsnC+AFdy7QTmp4IT5dFDHSp/VHi6QMS7Z4tqHyenDvxF70VDcp7omkdGU3P5FiAq3n84iL7heIvK+9jB9QN4hesLgASQoDsdjq1JAayCIEm7SsNGBqIQbsAk53m16wWgkld3xVm9LzgDdbh4W5xGlKw1Q6iaspzGVC54caxiV5RlNU/4uEZD2wXIqV/puEj5ZLdATErp+6tBqOtXtbHf0HPblqT9N39zOABETg/g5FZdm/FsKNUcEIzbLhj3mcd77PXjeRubbihHKqEMr5zKblvtjLKrEnI+kYMmGcL5BiplBXljMMw5juEVj5zDaZ3DV94u2J/8cCYAznBMnaup2Ya9eEwQUOWQ23bLuiVYRiZcp1yIV8YjSMi1sUeAa7tKf8kUAyKiCVcL+9fu7dRFndMleBWvZ0Ln7rnXC/bgwVz81E098u9CsSxnqc612G1weLYxkaUB4OI63ToQP+OProAr4Aq4Aq5ApIADLL8SXAFXwBVwBVwBV2BVBbbpZpeMopPni+FGt03WmgMKWt6t7mHrGTfjHFtp/4A48oLIaaJ0jSymGJbslrOEr40OMrRwM/37b03ai2/M2Z4dF+3v/ZVtKku8Pph6pf0CZxjAJFxGOM0IlgdCRU6TqPvfkEqwKM0j74qR1E36I4c75NCRS8eSAUacVyh3fp4Q9WpwUwFxWrU/SgwhPU33+AEchR0t+7YaeMK9gxPs6z+eUoc6daZU3hEOqpQgVp/C/FsFQPLzLbZva+QgA+ysNFhnXhlWzaWclA3yHN0g+Rka09KSDN3yAFwXtC4ACfpw3JK2SQjMqWJNZYqUtgkiaa24hgB9s61RiPzNlpuuNH+eQ9c1qi1Xe+u6nueaH5S7qTvXaft31O3phzpC0DrnktfelQONa1Yr1/6i0lGu691bU6EEMD4ImVbxwLGH9iU5lr75wtw1eBVvoMcZdTPE2dXb2Rk6TbLPhwRxyaCiHBFXJCCUz05F1+3MnMr2ZPPTlMJ5oKxx6XJu2mtga3ZS5cdRAeJ1L236FyDdg/tztnVLXkHzUWkzO+M62DqYEYjdBG3f9Gz8ja6AK+AKuAIfFQUcYH1UzpTP0xVwBVwBV8AV+DkooHtgO7w7Y7u2qhxJFVbtAjAEMf+8RwStbv4mF8BCR7vf/qen7LJKuPSrQuun7Ns/m7Z/8Tu77eEjndeBo+Z1AxQIvR6dKKu7oDLBlD8EdEGd5lI+yukePZJTgHktAARVkdleBbv3K4CcEYWmt9rTD3bq5wWVQgKXyLtq0zaJUPIFrAsOqyUABgiJg/3DTpa+rQaeAEvf+Om0SrMEr9iX4NXUrE6onDotAhxp4JOcbVmFeT+rQHnO+0oDd1UMr1j/cQX8v60OkABOOit+6pFe+/zHukOIP5ACCBO7iMiEIv+LMPXLV6vB/cQ2ZIaRBaZKxPA6jijWhmOouUnASvO5U58DHqXFeAflxBKuDNNkTYSm7x7JyHUnF5ucaPU6yVhy5cmVxfPtKj2NB8/H3R45H5Qbci0VSyuXOnI+/n/23jvIrvM883y7b76dcyMDRCCYQIqkKJKSJVlUtCTLsi1Zdnk81tg7ntmqSVtTG2rLu1O1u1WzO3/sTk2Vt8bj8tTuhB0nSY6yLFm2EiWRFMVMggBIZDQ6p5tv397n990+6ItGNzqDBPv9JHS499xzvu8553b3+fF5n5d8qT2DqZDpxbUANGx0swFV2U8ynpUrTi4sAS0gbLxLx5pP27efpdnDIjhjLlwKac2Ljp5bOYC9v/DhXvv9b4zYtNx3XAd0RiXfrEudN324Aq6AK+AKuAJLFXCAtVQR/94VcAVcAVfAFXAFblAAR0QmfcPDt/0DgJzf/qOr1+AVC+L2vSAg9b/+ziX7f/+3YyrdEwlYMnBtDasL3r/+L0N25pwygwQFKI366GPd9uCdrQFKNL4E4HTsQHpZIFMVQEjJErRf4OG993cIJM3b6GRJJWcq6+tT5zftN6e8rGE5qLoFROjch5umMQOLY1ESuLQcbkivyQlSjI+XbHSCsGx1bRRIAYZV9DmugPH5AFsASYTRUyK6Ar3iIFASDZw8lJZ+/QeTAn55hZKT4WUCWSV1liva3/3p/hAMHjnUeA1Zan1dKWWhxQVEUwIxmodet0/d7chXYm7MPwrBB64styb2dTuOsDZp0CJLXTYTVyfFnJxr6u4pIMy5P6jMNjSMBo8XdN4ZlO0WdA21Srs+dTCcnClGm137DBh95N720B00erARXkWPAdfoKkjw/9RCfhvX0/7+efvK30zbqzqH5K8xmE5S7q2ffHdbcHGFB7foA3Btn1xnX/hon51TJ0KsYHSEJEOtRZ06fbgCroAr4Aq4AksVcIC1VJEt+p7WxCMjI1u0t1u/G+Z/5cqVW3/gt8kRff1+/nfq9c9/ma9UKjv2/c/6y2rPtVPPfy2Ulakz2A76+U+I9I9emwzOq8ZfQWCDEXVHO3PuspxVNzpPxlSu9R++mpf7qBjyorjhLxar9id/M2RdKh3bJdAUj9XhA/uth9g328yNLCw4sJgHIyl8dmCgovJFlSM2VbTPip29OG8c79mX43Z8f1ywI267lPFF2DeOHlxjNCes6vvZ6bCbAJieerUSOh0WipSMNdukgBvACAdVToCOrCM8PrWqArjmsevoMYG5kZGrwR1V39P1H+FXE+pUOK75vHCyrPXnraJQeJm4TJFOmk/NvvnUpJ04NCc4oYB32YBG1a0wGrwex1lfe7P17FUmmF43V6sIsE2HY+KsK5KEvzBy0+tz/LF/RiMIqj+y+kd+99/K9/8RlWty7qSe1l6w0SV/NrIWrouy5ONnE0gnK5fSpx9P2G8N6VxKx2tDT37skYwd6J3V+3f22sOrfRFd2WjOz/5/+vmE/av/pJw0dTas6frAVffee9N2bF/ZJidHrJhnFpsbNCDI6ZoEHhd06XHtiudpqMRSmXmVQpNdvZHPbe6ga3g1535Obwqug504WD+/A/i8E0epREcEmmvUP+80DYrFon5uCpoXFkt6d5IGrD+fz1su1/AL6DYSIJFIWG/vzggOdIC1TRcmbZM7Ojq2ae/bv1vevLfz/DerkK/fz/9Ovf65SeIX+E5dP3+880fMO2H9lHrh5GHo75rg4FjtZ2N0A/9OWP9qa42ezwgUtaShPjeWZcUUztSl3+UdSwAWYGF4sqBucLMCMLqhB0otgKlypcneuJRQ18A23ZjTuRE4QY7Vynla7drfpEKr59g2WbPdSnRPpWqhY+B3npq1C1fKwfj0yrmqPftq3D7zwQ7LtmTtaG/iuvKwaE18fumsSsHExNKZhKnfn1VUstaUUJB4nK5vBJfXVOY3bwm5tjqUKRbXXLlOBrtqcnnd/O+XdEYAoly0MeVmkc2kKsQAV6LjU6J4brTJ7jrapu63XGN8AABAAElEQVR4CavMl/RzZS44ftCiT+Voe/vjQZfoNe1aP662pYNuflFXvqXPNX6Pm+e8SkBfeK2gudW0/4Q9em9LKOls3O5mX/PeZ7ydrv/OTuAViCsYlML8Du4127ur3X73T8cU0F5W2WGTfeETffaJx1pDiWHYaAMfuHFtFkj7t785aC+fLti43FltcmZ1qdS1U10n+bwRMNg4Fc4TQfRZve/K+nmbFeRlfdnQuVPXoK7DjtaFN1PjC2/B1/zuA968nc7/LVj2tUPMzs4GgLdT1z8zMxNAcXt7+zVNdtoXMZUJt7ZuTbOG2007/v4FArW0bH/zl+3QBvawU4YDrG0601xE2ezNOzRt06G3ZLcQ+Nt5/psVwdfv53+nXv/8Ar/df35t5v3Pf32/HdZPIPesut/hXOhouTGviOcrdBhbsFcQzJzi99Iqwcg4MPgDdidd//ym/sInBux///eXlAW0CFBwBj2s/KvBPiDIjX8YNgkENZMfJYeRqvCujSbpHFN2VWd7q3KDbnzdtQ2XfJERFKJcrFUOKWtOCirN2b//02E7f7nBDaGnhiaq9udPTltnR9L6ujMhO2op4AGE5QpFuWdU+6VREVDKlyu2R6V6QC2CwFlrb5fJJZVWCZscYlpwp/5u391T0poUon2TueeKZYXLq2RSOUVNKvmb1/Hw5aAegIMSwN6OVuUxtQhSmALFE3Jhle2iXD2EdO/TNZuQtYfywGjElb01rXI6Sg4pHcSZxPNtAmDMbbXxukLGv/zNWZVAVvUeblJwvJzEgmh/99N964IuuC9uh+v/Jx7K2k881BMA6dKy0dW0Wul5AD4Qu601a4+cyFpeP0P0kPQXYNLPjs3CK46LayydrgZn2clzM/bC6ZzyuJpCRtyJoxmLq1QyIzsWf4fd6sHa+R14O5z/7dCGn/9osFPXz3t/J59/APZO+/3f+D7CvJBKqbz7Nr5/b1zPO/lrB1jv5LPra3MFXAFXwBV4RynAH9cXFX790um8nbkkt4BMQ3co+PnDj7QrG2cx9JiOcAzuAXEA8ZnMoroL6NbfGL4dTgLOD+AIJW1Lb/g//liXnb9asj/6+ojgiWCfwMk9upn+F7+xZ1l4xXq6BZDo0DirIG7yr/T/cNOdSibs8Xs71uQaatQF6EJGFHlbLYJH37k6ba+/uXwpRwiDF4SifHA6V9Vc5MRqOK3si0B0Wa/CSAkQWE37Vij6sf0ZwYh4CEina+CESgHhc3GVH7ZnlEFl1ZBdtRLAyiuEfVwOGvK8j+1L21M69lV1NQwlhDpMXMfdPZCxxx+o/0c8rlnmeOp8MZQSci2SxXVJ1/HDd7eGnC8mCYRjflfGBE30GoLfp2YrdlUB+4fU8bJN1/dyXQln5ewaEdT7//5qRGsRgNX+EwKI5EedvZK3184W7a5Daw9v4zzeTmPptbxVc+c/5m9LgL4EpnTw9742Zt9/Sd0U63Ws9tq5YsgDe/+DHW8JvNoq3Xw/roAr4Aq4AturgAOs7dXX9+4KuAKugCvgCmyZApMCB0+9NGvffW7KRtRBDBfNj1+d0Y16yX71UwNyN9QURl5SXlI13ARyc9utrCE6j/Gvk1KxKPRmy2b19t8Rzp43L5cC4IjrznxApXdHBV+iAez4jc8O2mff327nL43asTv2hJBroBeABKhB4HWjY2igO26f/0iv/fs/G7UrwwJN2iaTidlnf7LbjuxXR71GohQdaJXP5EeReTQyWdFcq1ai1m+ZwXyaRZCADHxd1usa3V5Aov1yVr1+rp7lkdD66IpYUQ7WnoG0pbWWpB6rqDPhnsPJcG1wGBwIExNiXdifVhgEwgOlAFz7dqXt73yyz/74W+MqZStpnzGFs6ftn/zSLutVN7kplQXi4BlS98V6DlYd9M3pQTo/nhM0vPvQolsd2NqWbRbcqoZr/IpKAisCHH09afvU+zpDyHkE1pjD1bGqOjuW7NSFohxXgmjatlXrrFXkpJT9kDVeGS2sCWDx3iHXa0gusVy1qIwwfT1WsXYF6B8/mAmh+UudbitI9I59GM3JU8NuRyOBjbik+Dl0Uk0PfnxS8IqQLYZ2mc9X7TvPztgj97TVH/OProAr4Aq4Aq7AMgo4wFpGFH/IFXAFXAFXwBVYiwLcNJ+5UAiOlgeOZcMN81pet9Ft6Dr3rIDVRQKc5WLgpntOsGF4bNKSohYH5MZqa2lWNlHVJlSyxY0muTV7+rPKIjI5HKr6ftGptdF53E6vy+uG+8evzdrwuACHJs5N9+Sswop173znwUWIBZzoU/C6VWMqrYvLAVR3DgGIGKqaCxlNdGtjUFZ437EW+59+PWWvy100qxvwB+5sCfou5xQKL7rJB2DkjHKqiirbwknEfFtaNNfJhQk0vLYlA6BKXevW1/DUtS/vl4MM4HB+iGDq+QBhBvti4Ws2olR2j3Ki6PwXrTF6MQ60lUYMOrYwWgWb7tibtn/4uUGVgSnXaiBpXW0Et6usUNNualIpoGDVtMpd6wWGdf2BezpqgHTRvvjMtVwsz9n3np+yNy4WVQ5Yh4f5y3lBspp9UZ0N02qJxxQIAT93pWDPnczbhJxaFTm2AIBww9aWeHhtSm35etXVbrWB9rPSHtDJfl99MWdPPj+pfcppJvfagDon/tLH+wQm0zfVfLXj3M7Po+35ITpaVsK5xcV3UNdgSiBrPaOpaV4uu7KcoXJDch0swFJgbF5NEPiZ5sMVcAVcAVfAFVhJAQdYKynjj7sCroAr4Aq4AjdR4Os/nLJnXpkREKjfcD35wrT90id7rV8lVds1auoUNyxXCPCKDCzK3Th607y63317wp54VN3v5FahZAsXT03NtC7JkfXsa/mQMd7VmbJ77sjYT76785a2qQekbMStsRU6AoLG1AUvwkDBRaImU5fk3LljbzKAKG7Oh1WGNjVTlAOpaunWithfHcI0zoH8nky6HsrO4zjcegW9+LfZQW4ZcysBzuRcws300NEW+9sfKVi5wRHFMT/x3k6BsubgCCNEfjlnEHq/S0CNf42DcH9gD66zuMopi2V1LhS4ifRh/5TyrTSAVuNTTVZdmBPHTimMbZfKKQlcjwaQiRJFHFiZBiAWwT3mnRVgahw8Ny1IOyxnFSArGsCN4bGSnRbU6tH7iznOaM4vnFLQuOBVSdp1yil1aWROuU2mskQAXZMgX8ZOHF09kJjzD1DByTY+OWc/fHEywBRgGG+w0amSANqY/YOfHwyB99G8dtLn84L15JjhECzp+mluFszKV5RdtT7HFNdlryB6Wtc3P8diun4iBJbSY5xfH66AK+AKuAKuwEoKXP+Xw0pb+eOugCvgCrgCroArcE0BMoiefnlGLp3Fm+zJmbL95Xcn7Fc+2X9tu63+oq9LgeG6m6/IUYMjKKowAz6Mq+Ts+y/MClgohFiBy3R0y+tmk8fJacI5MzqlTKGRog2Plu0X5SjJKG9pOwc3u9ykkmkEnMC9BHzYikFGEvE5BIZHUGTpfilJI/eKkrXGASji8bIAnzBVKBUryPk0LygjQ1IAXmzf3XH9n0m8jmMSar3VA5cU54w56zDWp86H+f6kPfHuNnv+VE7OFznEumP28ce77KHjrSoNrTud6BK3nnJFArNxS0WDskgew4VUVnZU05zyswK5iba4/jMlYIO9qXBdhUwxaUGZHaV/SwcZSoS249IaE1AFzpGRxRXAtbB/Vz1kPnod1y1go6pzEw0cO1VlvRUEmF48lbd2ve6ocrzIvSrqBCa1fzV+VLfIeLgOCMIf7EvZ8Tuy9snH1MJvDQO9+ce4MqnA+5IujGjwuP5dHed9VBPAip7YOZ+5JscFryhrxaX2/Gs5vWfkYpQu7zmRt7/zU72hRHWtitwvqPrtZ5N2VmW9OLDITwOmcl0PqrzXhyvgCrgCroArsJIC1/9lttJW/rgr4Aq4Aq6AK+AKXFPg/NWibvgX4VX0xGV1PuNGGPfJdoy2bNweua/VzqikapnDh1ysNy6Vbf9AQg4ZSrJUaqXSqAKdv3RfiPMGd9Yzr02HjnT3H8/aYM/1EGGr5k1oPIHf0QCQkIkEeNmsG4v9ALCiQclbo/snehwO065yMsr9yg1QBITSLi0BN0AVyvYah26praj5zwnMUA4XDeZNePlWD0AfcCQn0Ej2EsfMCObQNbBbXfiOKINpsDtp9xzOhuyplFwrrZo/zqStGAArHFnzCnpfy8CFlU7pGhPYoAxsJYDIvsjmotwRYHVa5bazeXVaVFg9zQd6lgBC3jfArt196XoHPL2XgGTkbhE0j0PqVQXbn1NZ5EC3uhPqwYSIY1LOxIrAVVtrPJSFfuonOuQIS61lKWEbXGSxkpxA6CC416wg++iKCO9lzYtzQg7aThyA3bIoIe/pp16aCa7OCPg9+8pU6EL5xc/0WVbAci1jV2/S/v7P71LThDE7q3JoMVF77P42+5wy5Xy4Aq6AK+AKuAI3U8AB1s3U8edcAVfAFXAFXIFlFGiVcym6gWt8OiUWhJMGV9B2jV/4aLf9SCWBzykLq8EAFkrAgFqTM1Xd7NdUBqa8IwEJXGJNYbLNoWtdTM/N12I2IjcW29KVr69za/4cAFJRXsTq8wIKjHqFZR104LaghA0As9FR0E00WWDAjKKO9fXvT6krXsUeurPdPvX+Dmm/SHUAM3T1O3Ygo250hVAmRsFSp3KaACWAl6LmWZ/p9TPKZhasQg0PEyAOsNnKgUMN0JcWQCpXmuWKm1dGV13DpKaQ1blpatI56opZhwAN69kqFxvrqLvR1l/iyRzWM4/udsHXJQHddYdcXX3OBfsDEn7uoz32H/4U96BcPzPzAo0KjBcEAzjS8TAnvZ4uKLdKZbSUnXW1Ckgq64pg8ftUdrkeeIUGANAW6c77+uDuJnvlfMomVTbYpPngtqPc8ei+lPK9tuZ9wjFvp5EQYeJnGzlvOK+in328FYCoZy4VFKZfVkaYwtnWMICCh3an7Z/+8u7gbtvOn5drmI5v4gq4Aq6AK3AbKbAzfxPfRifIp+oKuAKugCvw9lPg3sMt9tfKwJoQOIkGN973Kch9q1wx0X6Xfm7NJuzXPztg/0IB1pQHhmqnBQJDedtMrqIcrBY5SHTjLyjCzWdMd5wBNqgcq6Z6HcrAojBy8o+2AmDhzsiFsO76jAENBDKjC3CC8jA64UU3v0vXtdbvJwWvCKN/XZ3nfudLI3Is1cPZv//8jH396TH7v/75YUsIPESDdR7UzXKHoNW4oF1KZXKDcoBEjq2knGloA0xpHMy1Vw4h1sCck4IcgDeAz0xOcEXnfl6v6e9JBajU+Nr1fI0DLBrMiU6BZEq9ealoaXXSAzDieGqRu2VS4e4drYvB89HrNvqZNdc7BdbnUBXsXFJtudFd3/R1aDglSIcbkOB6KAZrB9DxuV/uqr//uQF7TW6rF16nfJJMsCYryKU2Jg1+9GpeDjl12hRBAT7RUfLB4ykBlKwyr9YGUZZOEH3pLBmzhP3DL/TZH/3l1QB5E9IfMPPzT3Tf1Gm2dH/vtO8PC/i+di4fgBOEGnhFgDtwdV4XDeH36x3rAaDr3bdv7wq4Aq6AK/DOVMAB1jvzvPqqXAFXwBVwBbZZgV/+RL/9xZMTIVxad9Lq8pa1+460BACw3RDrXgWxf/qDvfYHXx+x0nRU7FS/twSIFES1dnckQ/7QHpVj0aUtdPsSTKKka9+ujMq6otLB9d94LpUW6DEusFBSqSKjrM/PqzvcpAAF4dp09+uj5Et3vZuFZZST0Vnv9746KgCy6AahZvKkgMfvfmXEfuPnrs8hyyqUPJtO2Z5lysoAIP3dKWUclS3gSGmUlUa9cj7hDGltyItibVcVov9n35uy8yrjpJNhpwKpP/OBLjsql9fGxvX6A/toDNCjDCy66OEY05TCIEQerbfq+mqEVxyAsHhKGbd7cA5nVCr5nLpDnlMJmQ4bSlnvU4nkgOAipYVkez18d0twWD3/+rTyl9QhUA6gH79WCAATDRICX2WFvZ+9Urb3PtBk9x5OBzi70fmHUkpByoO7EvbffXFvyHviXdUiNx5OrJ08euRwe98DHWpcIf1n5U7TRUk5rVirZQTV9w34LcVOvj587a6AK+AK3CoF/LfNrVLaj+MKuAKugCvwjlKgpytuv/DRXnWvq4RsKcADrgQcQjhJ+H67Bpk9P/W+LvuBQttncrPX5WEV1YXtytic3XVHXJ3qEnZAVToEuM8oX2liai5kB51QiDJQhxE5sTYzV5w0swISDEoov/79GZUaFUKHxGlpMTEzJ+giKBQ6jG1MF7KQxqeqdvZSKUDDcQENAtfBP+jOZyDdcydn5WLqW5f+7S1yOWXUvXFWJXtyp+0bTF+DRqwpGszhaypZfOP8bIAuOJguj1TtP/xF1f7JFwYUJJ66KVwCEJF3FUrnyN4SfKHsk3njPIrC0xNyiSUS0iwiVwsTAKJQgsnD2uTa9tH81vOZwPgANZe8iPngOFty6CVbbe7bitb+5PPT9tTL03JVzYVS2DfUYRAXz/tOUEpo4T0EJDm2P6XyTzn3lEtF58HZgjAjJ1v/mCfbVitVG5tQTlaD825zM6y/mnPiY1EBcst+7kNd9id/OxGC7uMxdeVUnevHHu0KP2sWt/SvXAFXwBVwBVyB7VHAAdb26Op7dQVcAVfAFdghClBWRllaNLipxl1Cbk+zbqgzMjpxI77VY3dv3HrVlS5+HmhUv6ePjjEhN1ROYdmH98aCY+cuhYDTUS90LhR0mRP5YU50dFsapB3tY62fWS+5VwzADFDt5LlZuXlqyiSK2ZTAA2BmarYsRxMuJb2AGqQ1DhxlYyr9e0Mlg2TwkA9F98BYc71bX7SraI9pASgC3hNLnFOrHQ5HD1BvVudzpdNV1lzOX607hoBRwDoGZZvPny7YA3cqj0k5Scu9Hp0mBfJKKkkke4yudgCpLoXa46ri3Az21IPZAaB04iPjKRociWQsyicZHAP4SBfBDQ3t4A++MWZPvTgrwGn23gfb7P0PRC6y9Z2j9R4fiPecHHoTApKcX0oKCyoTfPl0zh5UGS46k6cWQeD33Nuq55oFuwB3eW1fC9CS44bcea0lrvO21flk613XTtj+Aw912F1ymz7/et7Keg88eCyz7syxnaCTr9EVcAVcAVdgexRwgLU9uvpeXQFXwBVwBbZQAdqt49YYGq1ai1wRH3qo1Q7tyW7hETa2q0YHC4CCvKSRiXrQcZcylyhBo4vcgd2pazfjGzvSja/ixn+/SuKe1FM4kBZYyrUNT10s2GEFT3NsyrIi2AH4AKKQrbROxnNt341fsOanX8nZ0zo/o1r7zKwC22N18JJIqJOfMrcAMQWFbgNbIpdR4z5W+hogNqRw6NfPF0LeWEEOHeAG3fKO7M3Yj0/JMsXQmviXSiTt44+133AM3EZFvQ5gxBzQYCOD7LAIWtUaBEd/JsBzaEuZ5tLB44S1kxN24WpZHSPLoeSyTSWWx/ZllPFUB2hkMTEIIh/WNgV1SMSJBeSKXHM8z/VGmH2cUtFFfspTqw6ytP7H37qofKmZsB/0OH25YKfPtdgvfSQhfTYo0KpHrm8AkJwW9OP9A7xi8HFiunStWyRaRgCL/LR7jmStRZ0XX3lzVq43QURdYkBHlk5Hxvc90Crdt3fezNOHKfctYR95rMOlcAVcAVfAFXAFbrkCDrBuueR+QFfAFXAFXIH1KHB5uGJ/8d1xOyWIUaK+SeOlU3n71U/12v13tq5nV1u+bdQ9i3twXDUjKid87Ww+AJRyWe4m5caM1ipyh5hydbYufJuFADTecyJrf/5k3Car1eDC4vZdcVzKAaozneGxqrKB2gO8Ar5EXILcp60YuJC+++Np+5sfTtqUgATOmim5jAAgu/v1J4Z0AUQAonqUKdUhWLOeAfC5Kk3J9GI0K4Q+prIlFvL4A20BerwuUDc/JzdZR7P9zHu7FaSv3CS5caJxRSV+T74wKRdXPQj88J6Uvefetg1lJQFUDu5OKly8Go4djqFDtbYkbO9APVMMTZYbPE7QPQDu8nBpwV0FWJsLziH2W6nqxC0MYNsugUfgG4PS1AXWEzo7Xh4u25MvTlmp1KQMr7g98e42ab76NYaL7RkBx5ffyIU1hL3rA0Du6Vfy9nMfbFuYwfZ96tX7gnB03j9zWhRAkjyllK5RytIYUTkg5bIAPLoNss5f+nif/f43xgW7ygFiZQWvfvmn+u2uQ5ltB2/bp4jv2RVwBVwBV8AVcAXWooADrLWo5Nu4Aq6AK+AKvGUKvPpmLrRpL+lGPxqlcsX+/HuTdlCZLACBbKbJWtK3/lcaN9k4ekYmK3buSsm+pjldGC5al0rADim755jcRwMq9ZueEW3aFc1+az5z7AMKY3/38az98JVZywsecevPzX6nMp0IS9eXoXMdzh3gS+TC2poZWABWz742I9fMXABozCmubnFFgadSJS4gUQdpfd3JEAC9HvcVcyQUvqp9x6E5KqBLaN/VouCG9rtHwOhXP9Mr51LCxDYCBOpQOV53O2V4dYBFNteff2/MxiZVn7cwptU9MC5Q8ug9rRsCHp/7UI/9p6/O29mhorKblF2lAOv3CqZ1LOQlJVQ6utxAG9xGzEnRVxp1UMNUy7q2xyfn7O476vNufH0ESVlT5P66qODz3//GqEoqK6Hj4uWRZs2nYv/VT/fZYN8iBGvcT/T1jEpLX1RnP4LiGRwxzEQfKL08f2XOHjoRntq2D8DMI+ocePI8pZ46p4J0QM/D+1vUqTIe3lON627T9Uy3QgDXoT0Z+29/ZY9d1Pusqo6Ndx1KymUYlT5u25R9x+8ABShLxQHJ7wwgPt09vez0HXBifQmugCuwoxS49X/t7yh5fbGugCvgCrgCm1WAYOeou13jvigjIlC7vUUOjeaY7elP2j3qznerR0ZwaHqmbP/371+1K6MlmuGp+1/Z3hTQqumGqaMdB1QdVmzl3HCmcBP2xZ8ZkDMtJlfajFVUV9WicqtdvSnrUAljewv5VPWyOwBKLkAAMp5uBCUbmRsApyBwxu7oRsZICUg0CTRlBBu61X2wRyWU73uwwyiVW+/IqBSvWZayVJJMpJrWQjfBemkZoOoOld7RMXClcUbB4I3wiu0KpZqdOV+yh4+3XpvzSq9f7vEWAZZf/9mBUAY4pByrPnULbMnUSyOBmZyX5QbnqiAH1t8+k7MrcmDR5bCjvTmcL+BUKkDGldeCFgTlcxP+A+VWAa84j7yWUrzJqZJ957kZ+9xHupc7/LXHcHHtHaRMUOBK12rjldks+NbdeW3TVb/guHQyJKS/VRAZx+FagAA6PfFoh4Lqm1R2Wglz6BfkvFdlgrv7U0GTxoPjHuyRw45sMLKyCLKnjA2nXVRy2bi9f+0KLFWA982wSpyv6PfGX+r3xmsqRW3Ve/m//txeO3EktXRz/94VcAVcAVfgbaqAA6y36YnxabkCroAr4ArUFRjsk5Mo+EQWFSnLUBPcPdyBKwWnJnBzYaigG98mlXitXEZFKRsQgSovTD24PIJrSHBno4M8n9/+8qicV2r/tzCAWHTJe+rlnD0gUELJ1HYMuqSlkin7H/7eoH3lW2l7+fVZQYFma28T6FGAenen8pEEJaKcoZroBYHv8S367Z8VqAJMTclVBLjhOLi9MgIUhw9k7NAuudD0mVI4SujWAjcadcIxRvh8HbPELSmHEA6z/SrHvPNA+prTqvE1jV9PC67MqfqwqnpLQA0IDfcV18DGzzjZS7jflC+mf6wdTdF7pQwmrrevfmfc/uX/c8WKyrSKxoGBlH3qg13W15W0vX3Jm+oDwGIA5HLqKIkOXLvR4NxOCqSuNjK6Vz+mEO7dvWm7eFV2toVB6emx/Wk52BLhPK4GOQEC31Dp6PfUTfCKyhl5J95/tMU+/7Ee7Xt1IEAQOPBxaFQASy/u1HlGz2id0bwaP5P11eadARsl8a/XqAA5fFfGivZv/vOQvXm5qN8B/EQo2T/6P07ZP/nFQfvZn+xd4558M1fAFXAFXIG3UoEt+hP2rVyCH9sVcAVcAVfgnazAI3e32fd+PKMMrFxYJjALB8egSvOyKgFpHCOTVQGsxUe4R8nJsTGTU5i44BU35eRDkUVEODbOjjZBEuAL7o71DIAMeU/f/fGUnVSZ43JjQrlYPYJIg3KLbNegrIp/n/9wj70kWHRJDoNYABsqJZQLC1DHOhmRY2er5pKR2+vD7+my3/vaqKCHQISOQ+dFIOL9RzKhHIxOczh07lD2FB36snIrrQdk3bE3HVw2ed2ANgkccZ565HoCIq02cARBqgBIDJK05uVI6+tOBHgZHtzgB64tMs+4jqjGI/+L4H6g29JByd+/+o/Xwyu2OaeOhlw/P/dEr2Dj8t0LG/cF3MGpRNbV0NgiMGUb5KBUc7XRLv126V79n/3ybvudLw2pFK8crpGH7m61L366y2amJ3UuuVZuvqfnTubsr74/YW/I5UYJKeNqCJ2fty9+ZiB0VLz5HkzAKm37BPE4P40wbrXX+fOuwHoVKMp5+dQL+VD6S+5aNOYExX/3y1ftpx7viR7yz66AK+AKuAJvYwXW99f623ghPjVXwBVwBVyBd64C/+gLu+yr35uw0xdLltFvrrTyrjpbbwQhISppQQbcVqcvFOw//tmIvfhGXrlM6p7Vm7AvfKRHOTopAZB65g85OgxKy5Lr+K04k6N0qmrnL6/seqH8jRun07rJb1e5yoDAyWpgYGH66/4E3DgqVxKlVZRZsR5Krbo7FJgtpxmDkivWvZ6Bq2o5Nw6OBgAex/3E41322oWSSifnba9KOSmXa1LgOsH2bJcQ1Dqr76sqD+tQuSMun7XqANigPJTzyWvWA78AZnff0RKC9UsKnsK1hB6Pnth8UPl5QSlKRgvK6YIXJuV0m8kp+0xdHykrbRzPvVpQ+eai86rxufOXKyqNrOn6FAVbQ08CYOQHHmpX5hrlkXWIhS6D3Wn7iQdXXxf69es67FIJ5v/8G/sF32ohYwyIWyrpHIq/rQYHyaw6ea4QcqiAV4vdOGv2gjpDnhtqF2SkRLN+3TWud+nXHOt6tZZu4d+7AptXgJ9Bpy6T/bYIr8JedYlSpk5ThF5vrLh5oX0ProAr4ApsswLr+FN9m2fiu3cFXAFXwBVwBVZQAIjx6fd3h/wfuu+RifUDdWCrl4HUX4Qza09DFzZKk/7dl4ft6RenrSjGxG3L+HTF/uXvltS1rNN+4uGu4FrBSUWRIsG+6wFYs8qTuqSbnmK5piyVmE2ru9vS0a1Q8Rk93pJrDjf5gAYg1naNTgEb4FGpIgqhAHlgBOCHAfigvG+tgxIxMpd4PZCBcq8IzOA2ojyPtTH6euIBnLXIfUQe1LlLJRtWsD3Oq5pOWJPK62a0LwKU2UdJmqW1v8bBeeV5crVwE+HuauwmuBGHDlo8cLxFQfpJm5gqq7QxFoL/gZ+bGePTVeXpVK7pE0pcVbkKrBvTNbY3vVhCh9tvQmWGK4151c8BGkflHhzUPJcO4BDXGPpQ8or7EKD3a5/ptydVvjemUtV+6f/Ew53rKq8DLuEMnBKIjQAUcKtV52+1wRVVUefLovgZ759rQ19XFEg/IicWjwtTXnvKv3AF3koFugTN9/ek7emm2fovA02Gq5N/Kf2Xj9Dd9K2coB/bFXAFXAFXYE0KOMBak0y+kSvgCrgCrsDbQYHI0ZFS9tL9x1rtNblAikXlLwlM3CHny0B3/dcaFSJvXCrYq2dy1+BVNP+ibrC/8u2pEOB74s7W4Ehie6AJ0AZ3yXID6MLz3PHMy0xDSPtF/ePGh/wiMq+KpcXXtqm8EUcUTq3WrFxIcrjMCGQQ3r0eF9Fyc7nZYzii+LeZAXwAUAEIAVg1OYQIUW+O1YOzASp0rGscuIhiJcQhZ0pIR1+mdZ7mavUMLnKnyIpC34UGeNdeXhSQJFwZpxfnFlhVma0a5W6NEOvaC9b4BcHyLQJmR/aqTE3XB+4zjhFdR2vczXWbcQ3gqmMsVM5pyXXIFPar9bNNdAyccO9Wx0Nyx8R8bhj7VEYHOG0sa4o24jxMBgi4cF0JGHEdUka5S9fczz7RE9YTbb/ezwCxHmVPhetaL57TgsbGVt8LYPSuQ206T1MCVgqRD3i4DgOyAo90ZNwIcFz9yL6FK7AxBWK6Zn/jcwP21z+aUsODSrhY9SMq/Ey4T51cOwT/bX5lN+3GjuqvcgVcAVfAFdhqBfTT2ocr4Aq4Aq6AK3D7KTAg9wj/ACwAIYBJNPh6VuCooLKxRaQUPSunj5xDuFfOX1VmloKj9w4mrV3OE+DJyJgAQW815BlFN+E4awAwjLwyj06dLeq1JXVQK9voFGBHrhi5Z0anFWYteNHdFVdJVyLcHI3KiZTN0BFQd1AaIXd+c3wp7Gc7PwBJgFc4sABTDFxGgD4C2euiLgqOuwowFc6BNsKhVdN5YQtKEEP+2HxMmU9Fm5yNy6lUC+eOsrVZ5ZONKJicYwGXcHB1CtDg0KIUMhW1NwyzWN8H8qjoksfxI3jF/jcDECNwB5wL61uYEueVwHzWtKhMff29nTH7uQ902u99c3JBu/qLurTOB45lgjNscJn8qjw6CmI1DpyCnJ96mWbjkRq3Wt/XEWwj8H4tg/N0/7Gs3Xs0az9+bVbz4dyZwFrCDu/PhG6C0XtnLfvzbVyBW6EAcPUP//djCm5/M7hnKSt88O4O+7Wf6Q1QeGbmVszCj+EKuAKugCuwGQUcYG1GPX+tK+AKuAKuwFuuwEo3ykfU5YySMcDI0oHbaqagtuoKwp6W0+fUuZw622UFmZSLJWiSbi1aSQ4X8pxwC0XOKhwxp8+X7BWFtlO6VRLUmslVBLUE0eTE6uuUy0f7zmYIpRal0vaAr0kFmRO+3a8bKFwvb/cBqMI5FMEr5ovLiHK44C4SpAF6UJoJTKnDq7pzCgg4K0A4pJb1lASWpHM8Log1X7VsIabt5+38UFFgig6DMUvoHHUI8IVjLMCvRKwOaJawm7DNej4AqghWLwn6AJhwwa10vay2X0ApbjrcV9OUBKpEEyfUNNeX9AJo4QCkVLPxHCcEtYBen/t4v91xIGt/9cMpG1WJHdfWUX1/p/7t35UMwfJL58AxlxtLHWzLbbPdj2lZ9osf6bX+rmZlvJVN8We2f0/aPvqeTuVr3VgKud3z8f27AmtRIK1y5n/3m0fCz62ifi5QkrsZl+dajunbuAKugCvgCmydAg6wtk5L35Mr4Aq4Aq7A20iBoyobe+KxLvvSNyYEEBYhFuVaOEUyunEBNlVqczYrF9WFYblj9H25pDDfkyP2+P0KxD7RoYDueskYDp6rE3P29MtTNiF4VRa8mSlUBHFUYofbSBYUzEKpeuM7OcPU6VDbNAlwpJPKi5ITqOOOOqi5mUw4bABFvB7Ygotoo9DlZse52XN0NYzK4xq3w13E/MhJwhlUk6sKQMfaW1Uyibvp0nBFHfIqcqgV7eLlkuWlAVCrTZCqu2NeXesKFhP8SpPJFZxpcTu8Lx0ytoBkOLYqWjvwZrOlkMyduQGXNjumBa9wckU5VKKT1q5aOtbN5dWrcPh2ZWu1Za//04rnp2bnlT8Ws8dPtNt9h5XU3iRAlxLskg6skTJHrsWlg26Sy0Gs9QbxL93vVnzPtT0h+Lu3P2N7+lICuHQGXR7EbcXxfB+uwFYqgIMx5Zx1KyX1fbkCroArcEsUuP6vrFtySD+IK+AKuAKugCuw/QoALv6bX9ptu9V58D9/bcSmFbyN6wfQ0Nled8k067cg8GlUYAq3UFLwqVlWnau1kv3VDwSc1Dnw7jsyIZx7RI6iMWWn4NyKusYBseguh1OovU37FHCYlzMnof3k1dkK/0yvbpRwblFK+Pr5op04kr3OodOoBO4mQAkQh8H3dPKja9+thBZAmpa0cqgUFVOfCWCuObgVJKvm0mQExrcKQLUJsOXlWsNxhtPojADVs6/mg0ttdHIRHHaK2+xWyH4mVQ8iL8qVRa4WgCZ+pcnuOpgWP5R++h/75wazZcGZ1ajRdn9NThcAkXUC7HBxMcfG7mXMC2dVVee1rzMRdFnp/PD6bnX8w6mGlv3dglVrLCEldL8iYNjYrIBrDNfIZseE3g9l6U93zI2AwqtqkkCIP9lec9V6eSmB8NcumM1O0F/vCrgCroAr4Aq4Aq7AEgUcYC0RxL91BVwBV8AVeOcoAIT4xY/12Rc+2mdnFbh+6mzOLo1U9HVRDio5nARlSir/q8zVu6YFC5aWD5ACxpwWjBkQADurrnpDY2U5qhRYLbg0pWDtsixKwAXxKpWnqYQw/E9B5XptTGQsnaqpk16TMrIq1iI4VlRO0MxsxSi9OqKcoOU6AlK2BxCbmKmErnNtcop1tDTrtbVlt9/OM0XXLgblagAXQEyzBG3sHgjooowOvQA8lNjhvHr5zRkbmwTXLI5JNf/Kq23d7j6tqV3QqnlOayJPiwB09DSV+wF36t3xCBe/1YPywEllcmV1vhist701FmDW0rkAt1J6frnzuHRbvgfIrXcAqzpUAkkJKxCU8HScXJsZALpX3sgLytYDr1jjbpUz7htY7Jy4lv1Pq3QW4IjrMMrQquliKQW3Y/3aWct+fBtXwBVwBVwBV8AVcAXWqsCt/+twrTPz7VwBV8AVcAVcgS1SQPfodkhd6Pg3MlGxbz0zZeeGSsFJlU4qB0U2krkEcKnuJInpBTJryW01Z8+8krNzlwth+0lyrwStcGrFRHUqgjY4k7ItlILVgUe2JWFZwYpiuf68GIQ1a3vmUJLjZVjH75QLbLdsWo2lgbitLg6X7KQC4il5BAplEiU5dhJ24k6Fc93iAaChM1cIEhdkYq44riTRdYN1dQry5AV/yH+6MlJVR8br4VX0grLWyPpj6maI80f8y5rEdToF6tg/zqZBBfNTWnirx9hk1Z49OWPffSpv+coFXStpe/i+NrtDuU69cllRztfowmJ+aLTWAYQMbjMJ2AgBV3t90D279uOstr9zArkRvGJbgOxllX32KLdq6bm92b6Alxk5xAqCa/NyLQKzgHQZlUb6cAVcAVfAFXAFXAFXYDsU8L8ytkNV36cr4Aq4Aq7A21aBvq6E/cyHegIouiiIJc+UbtxjduFKwYoCR3RiS+vGPCmC1SR71QU5is5d1j/d+BdVUofBRJxGN+9mrXLDgBbIqWrJJAVjaoJhzdYt4EF5HOHtlGlxc1+TkwmKNaXcoKujZWVCNSvsuv5rWAwhdMsbVnYUDhZK8ijFqsgFBiTDIQM8utWDUrXVytXqmVX10Pqs1qSV1jstrjRZrYvyu7SAX6HUZDitBhV+TqB7i3R8K+DVZbnyvv/CtP32Hw7buNx18t/Z3z47a3/y3Sn7e5/us88+0RO6SOIwizKpgFdL865WWjKdEKPuhWyD044STODfrR6UqC4dNQGoCbkDexT7ttbRLQjLOhIE9Ndw5+EyUyC/YOZbMQDLdK0E/NJdDhi61lLNt2K+fkxXwBVwBVwBV8AVWL8CDrDWr5m/whVwBVwBV+A2VwBXyz2HM3Z0fzp0lbv3cNW+/Dfj6o5XUAlfOXRR6+1UyLbCv6+OzdmVcYElgS1ujgNYEgOIKf8qJudVr+BLmo567WZtLakAmpLKeZoVKCgWq6EED3hFVlDHQsA3N9bkLGVUZogbB7gBCCIYfR6+wHE0gCWUZ5GnRYbSRsrQ6nvano/Mb1TOJeZN08XJ6ZrtG1QLxpfyEmv5Y7arNBGXFdlZbYId/SrRbFeZHCyHf+gSleXlBCQmtP+ENOoXFNwO4EPZ4NXxkv3RN8YW4NXivMeUW/blvx6z++9stTuV0QVwpJSPeVLquJbB/hvhFa9BN+BPtM617GertuE8LTeiMsDlnlvuMc4hgf508eR9QXOEvSpDXO9+ltv3eh8jk25akDAaOOVwNHZvQxkq540y5Klp/ZwQxNs3kAxANjq2f3YFXAFXwBVwBVyB7VPAAdb2aet7dgVcAVfAFXibK4CLBqcVOUf/6Au77MVTs/by60N2z7FelQY22akLJXvtzYIATZ3GcO9f0wc1Jgy0hU9AiBY5sfq7UyFsvU2ZVbPKsUoIcAALcoJPZWU8sR0AhjI73FQcm5thABaB8wzmkkhWxa/qlIGytbRgGKHwQI/1pRSFXW7bB6AVzqW8IB2DLCVCzbs7Y7ZfUOq8ytKWDrp+7emN26MnWlWOJy21fbv0ogyzXVlayADsQauT55SldXo27JP99HQk7b0PtK3qCFt6zJW+55RSPgfoGFOI/6XR0rKbXhwvC2wWA8BiA7LA1jM4b8sNHHpSLTzFNiV1s+R/aLGdEGiXOgaePq9GAfpfNLjuetpjcvwtQqDouZU+M8dDu5M6//FwbXLOuLbfikEnTBxYQN45nc8mnaNOQVEe4328VSOnBg7f/tG0fi7kQ24bzkqywz4rR+fbDS5v1Zp9P66AK+AKuAKuwNtJAQdYb6ez4XNxBVwBV8AVeMsUAJ7gymqXleTgwZbgLinqBrglraymGNlPKk/SNvzibFaVFP8I1N7VkwplcD1yCHWpLIzSJeCUMt7D6yYmCnbmSkU31nN25x1ZG+hJCtwQji7XkgAVgyBsXFm7+pIKfS9bTXCAkVSpXZsytToFd6JtwxNv4YeqAMG4OiOOqisjnRVxUhFWDwwCIHQpzwq3UlEgalhgKBrkJR0YSNj9x1uDbrjgKPVqVRZWf7fKL+tLDoCIrLGXBBPnqK1cGGPS5cXX8/bue9XOcBMDd87r5wohXB/o1qJ5BcgUJrB4vMVD1NfGVKI5Lj63+lcrObXo2sgAvpALxryAQgBNriGA0HaMXgHGWi0t+FgK1ygw9aCy4WKaD10n1zNYG+WzKw2C+YfV/ACHFPlyXXI1tt9k+5X2s9rjOWk4rRLIV98o2JMvzoRmAt0Cnp//aK89eGd2tZev+flX3szZqfM5NXOoA20gICXG3/3xjD3xno4178c3dAVcAVfAFXAFXIGNKeAAa2O6+atcAVfAFXAF3uEKAKGOH8rY+x9qt6sCDOVyUXlU9RtX2AMB53cJeB1RGSIgKwR9wxwEOoAzoxNl+/6L0zY6Xg1h8ECS3CtzchDVKwQTiZTtkhuJgQOLbn4xZWjdfTBjZ4fKwZvTpwB3gBf7Wy2LKuxomz/gVrqC60r1lNO5quBLNcCJrvm4XGVy4AjAAHmO39GiHLGYDas0ryxnEcaefq0DPQF+aWmbIS9LG5OBFYEhPmcEcM5czF8Hr6Jljaij42bH84JgwxOLbqtcQeBGMHGXSuLO5Bcfj47T35Wyw3sz1+YYPb7Wz2mVoRa02wUTX3gZ68RZxzVxcZgOmPXrCuBD0H14Xq9bCX6t9dgrbUdjAP5t5wDQnlduXHXhPcOZK40oD643KSi7MvTayJyqAs1n1DH0j781rmtToFly8v77rd8r22/+/f12YFBvuk0OlkFJKWC2cWiZgoFFlSLLHahr14cr4Aq4Aq6AK+AKbJ8CDrC2T1vfsyvgCrgCrsBtrgDlXB9/vDO4Rv7yB9N27lIulLR1yBV1z9EWO34gJfCkjoJyFjXJnjUroEOWVVOxyU5fLNnEdEWgpxaADw4eugv+6OUZ6xW0mc4lBEYWb3iBVJQ9dbRm7Y59meDMAQIQeL7UjUOuEuWH3Dzj3mrNxG9JYHVOXRmLAgQM3Esp3bAzl7we5/tcvl7+mFWt4/5d6t7XlQyB9QCa3k65fOTCoVSwT4HtdB9kzTh/GDjWgFeUISZY1DKjeZOp3ASYjwhsNA4CzJU9br/yyT77N78/FIBj9Dzusk882q7S0I0DFyBUV1tCpXZqECD2gQY41vg8qesjglfRMSktLFeaw7abXG60y2U/l3ER6ZwhdVTCuuyGG3yQ4PoIXkW7wKU3qce3EmDxvgIGPvOS8uuk8TxvCg3Jq26LFfuSss3+2S/viqaw4c/ana5PPt44MnKxNWItpkApLO8JQLgPV8AVcAVcAVfAFdgaBZb/C3Fr9u17cQVcAVfAFXAFbnsFKN1734PtdrfcVkPKScoLTrULaLQKQrQqlJ0StLicR7hphsfruU7BZSVoU6kqgF22GsruuMGdlw1lQl3uxmcU4K4Of7q/vWHwGGBnpZt8gNGMMraAWxU5T7hdx+XUq+6K2z1Yo+Lmw2G4MZ9TIBg5QFU5h0bkakqLBA10x1TOJSjSLMeZ4F1S6wTA9SvsXpWYNj5ZViB7WpopvF7zLubKIUOsTaWEwJuK9tWabbJsKhbcNNGaOBKZS5sZdHVkvo35T+wPiHbv4az997/Sb3/xnTHLlZPWrUrFd9/bYXvUAbKgeW5mAKKWK7WTnAFyAHYaBw6i9cArnHHsghLE5a6pxn3TSfHrT02qOUHVdJnZfoWQf/Dd7YJAG4d0jfuPvmYNy40ai97CwXrJ3iqEcxuYnD7UQSHXzIRce1y3vKc2Mzgfdx5otRfPFAVqF52A5N8dGEwHVyENCGYCsJ4L2Vu8R2sCs22tzYJfKrV1h9ZmToG/1hVwBVwBV8AVCFEeLoMr4Aq4Aq6AK+AK3EQBbpLJuOLfSoMb5OAg0s1ys+DNYF/aXlYmD84sbmRDJZXu3YENryug/GOPdV7bFYCL2/q1hHdzk8w+CIqPwAc5SinBIwLkt3IU5dJhbriIKIXDDUaWES6pqFyKtRYVVE9WV1ZlgYALVkNGGK6jTmUREXCdlXvlqro5TkxVQ74YJYezcm7hyAE49HUm7S6VGHYoUB/I9Mg9HfbSGzkBv7IcUk12RK60Ow+ow+EmBqAD91OueP1OyJwaUPB8rpCyD7xL8Kq7NziTgG+cE8Lmt2OQg0X56WxB1K5htEibtYTFc01NqQsg+VmAzYo+t6s8D1i2XMkp7sD/+NURu6jcJso4cZiduijAqsc/+b6uhhls/kvcSjiQoms02iNZaFs9WO+uvoSdvaQ9LxA8PuHyGxSA3Kox0BO3T0mnb/1oKkDkbNrsroOtds+RrL6vKdi9ZpPKh+P6n5rRNaNuDwDZ186WdY5r1iPIfPehFuvv8v9+vFXnxPfjCrgCroArsLMU8N+gO+t8+2pdAVfAFXAFtlEBuhnO6kZ2XmVpDxxrsedO5tThrl6yhv+jWcACt9ZMXoHkp/NyKyVC97m8bm6nKL8TKCHQvFslhisBDIAApXyNYIAb5mm9vjW7GIbeuEwcKMCE9bh6KLcjlD0ahRKOM0LlYws36XXnS3sLkKLeJZFtqQis53nNyY2iQHutNyFYgvMnX9INvfZJaPkrb5bDDX/TPEHqKr/U/MuqI3z33W0BBLZqvx+SMwggFmAEAq5jAJ0APKw50hLAdnB3WuWdhWtrS6mG7p7DLcGhQy7U0LBKOZeUDAK4tmMAqoCRTU3xkCWGi4qQ8672tf15NqMcMuBVTtdPUZ0uGeNTuhq0H3RrLD0tykX2Z9+ZsB+9Oh1C2Cixa0kn1EUvKaCloPWGc70Va+XYnVrHxHS9rJZ90hChbxucgkDGX/tUv72hrqEj48XQhZDrcECdQT/53u5Nu68iPbgOCbw/uLs/wFnedxmBYwB1Tu/7OoiW60r6A5i55r/65IS98kbRmjXHgY5me0hg9pPv7bSjB0S/fLgCroAr4Aq4Aq7AuhRY219I69qlb+wKuAKugCvgCuxMBQAlOIjm58myitsXP91n//YPKzZElzlVHcXlAEoJ5qTiMXv+1Rl7U06sO1WaOCMXDSV1ADBgRLmqgHeFni8HnHBBccPcOGA73LDzeGOpFOAKEAXkYOCEahOEWinzCIgxW1Ap5JhC61XOR5lgTYCJfePwGVYgfZNcJXN101RwYQGFZnUMgFEUOs4cyAajbOrquELKgSxyaaX0V0elrHIvOY64uaeLIQCH53Fwjcqh1dOeDN0go7VHn8MC1vgB5xdONUAOA2ccLh3Ozy6FiDNnXFjyi4WQ/Mj5FqBLW/M15xA64VLq1HnZzOA8lFXuCQAhVy1aEzp1CZYVBQe5ZoCbyzmnljs2a0M7LoVG+ATY5Hi48iKAxbl7WR30zpxTOL6e43rRIi1XqqgbY5PtEWwCeAG1tmqgXbcaHZiuH+bXqa+32iHYONd2wbL/5R/uta/87ZQNjRXCef3AQyoBVT7ddgycc9EgeJ/B+WXMiZxeUbD773x5NHTsrD9qyjxT18nSlLUJ2N6xNx1KaqPn/LMr4Aq4Aq6AK+AKrK6AA6zVNfItXAFXwBVwBVyBdSkQQIVK7o4fytpHH+ux516fEayZC137CCsfEai5oE6DdDX89vPTduf+lO3fkxaQiAkiJJQhFRcAmlM4++JNcjQBgMqoIBZ5UQzumXEIAWc4LmV5hKpXBMwICKfULwJWgAQgBTAHtwglZ7wGeHFJ3fBGxgs2LkiVrynrS/sAthG6TtbWqEr/AGSU/gGfgGH71d2N47K/4QkBL4EuBiWAwLhuwSiylQhOJ9Q7o311tKpD46S4Ro2Ae+1TJVZzgi64pbAGvSLQQrfCtYIcXhUNINoFuYmGRuuuN3TB8UPgPbAILYBEPR0qB+2IXnX9524BrL7+lIBTPSCfdUZg4vot1/bdC6cK9oOXpgPUOCT3zv1HW0NeWQTN0G8jDi/mFM6dziH/Y1AiWJTLrVarBlDWqbWQ3/XGpZI99cJscPmxLoBkTC/hWgD0oUdWWhWLenKLxqSy3l46k7eXT+fC9XfXkbSdONyq83/jNb3cIYFwEZjjHDSC2eW257HO9oT96k/3BmcfukYar7T9Vj2OjoBEPpNPd2GoaL/9pWGVx9bfo9FxZMy0SyMlOykH4OSMYO1NSpKj1/hnV8AVcAVcAVfAFVhUwAHWohb+lSvgCrgCroArsKUKcAP94fd0BkDz8inynHQDq/yhGSCDoBDgKK9k91MX5gNAgP9k1VGwpidw7MynAUy6K14YgBgeB2wVy3VnFECGm/ULQwX73a/k7M0rheAsGpTTaI9AzH4FTO8TaCLDCoBDTg/7xKXEoNRpSI4rYNaU8numBZoScowB13BQJRO4uKoqTSsrXykRHDy4fHD1TBBGrzkSdD8oxxiOLVxBKR2rTQH3TP3OAwq53pUSSEqGMrcLV0oCNnKR6fAyANVhirYLMEYUgK6NJXU6zBWaw3xZ21rH2csle/NyUeHwch8JqjHPOVGD3b0pHYvcrrXtCWCS2oLopG89M2Nf/taoVRWE1CTb1ZsXisHF9gnlKPVtAbwA8nGtUB7K+XszdL6kvFTlbHp84M20tQtiDY1WQoc+gBAAaWpa14DO67zcUR2tSfuIOm1u5QCK/Ze/GLJv/khdO3WiZTi0Z15J2mc+WLMnHukI5/Vmx+NcUR6p0xdGQe8XyldXg5pA0stqtABwJVx9/65kyG272bE2+hyAjSszp7VSJnl1TAB4smqvnMnZH8sFVlKo/HIDgF2SGzGp94gPV8AVcAVcAVfAFVifAg6w1qeXb+0KuAKugCvgCqxLAcrVPv5YR4AMz6hssDYvN4kpX0qlePP6V1PIOze7dCckewmwg0smhLTrHhgIRtkfIdCRI4X7ejrnJVUdRcbUa2eL9lt/cFVOp4qV1BJQSVg2Ng0E0msqcypZmw/d9Ch1AizNz9fhFTfhUdYVYIvBvgsCSJQD4mgCSJFNBeygs2BIxtY2vDavUsBKRXldBHZr+xDADolaMoBnBwTRzl8tBVA3W6xal3K+WsdioVSRYHicXS1aZ03H0tPBGYSbqEOlZwCQmw3miePlGz+c0nznQobUA8ezdvxANsA63F3LTOtmu9z0c8z9z58cU3ll5GqaE8ia07kq2EPHW6xXrqfNzgnnGxSloLLMV17N2WvnCnZZDh+C8mNxnet0TrAoFhxfhLsDO1O63naprI6yyj7lFsCrCAAAQABJREFUrf3sB7qth1K/LRq4355+cdr++DvTujYq1iS4mdT5rc6V7Gs/GLdH72tdFWA1wiumxfml3LRTZamxmFyFyTocbZwyx/32c9OCc2TBxQRtUzYrSHuf8s1wqK0HhDbud+nXvB8AusBn3l8FZY+RXcf3l+S8+pNvT64Ir6J9HdufDu/D6Hv/7Aq4Aq6AK+AKuAJrU2Dr/mJZ2/F8K1fAFXAFXAFXYMcpkFF49U8qkJyA97OXigI0Aj4iDwAsyrkouRMf0tdy/sglpFzx4DahJAzH1YQ6mnETz9eUCPI4rzE5sfLa5+986bJdHCoFqASAqmq/VZXtpQWWBtTxjNB3gBO/9Ck9pPsdg31y4w1IaczVotQprZyerOZNGWFbVs4rgTHKuHhpudIU3FTAuRnKFUWIetVFsNEtFg7Q8AH31+E9qQAh3rwk8KJjzCi4fUqAJ5mIyTHTLBCg4O2elGUX3Ck4qAh3Jz/pZuP//E+X7ZtPy/VSrnfzy2mfBVGwuNrQvUsgi3Wu5t652f438twbF4sK9o7gVX0P4lc6fxWteW7T8CqaE248SjqBoJPq2Dgzq7w1rbeijK+y4AqZZXWX3ryccBkbkVOooufv3J+1n3iwzfYK9GzVwAE1odK4b70wo2NXQlldk85hKYDPeRuTQwnHUs9NDghs1EuuDb4fVo7aS2p6UO/AOSe3X9oeurslgDDAFGD3P3911J58bjJ0tdTy1PEvab/ysT51v1SGWZs6W+oaB2xt5jpgXqwxmh/HndH3HG98umJ/++x0cCZem/wyXwyopPXvfLp/mWf8IVfAFXAFXAFXwBVYTYGb/0W42qv9eVfAFXAFXAFXwBVYkwLcSH/goTZlPBXs5TOzJiOWqFGTStzUtU/wZn9/Up3JsravLyl3Uv2GO9pxkXI/3YCHDodR3pE+Uy538s28XRUYAFDh3tLuQjg0JWUh70jEiZt34BIB4lmBJLZh4JqK6y8BSv/ScyovVFkgI6nStIRsTzjCOpVbBTDjpr2mfVKeFVeAEllcHS3xkPsDfUsIbq02cI31dzery2KrDfYldOxmO6+SQsoZcX1lVd62RzpE82N/AIybjSsqGXtW3R5rtbpG4cWayoTK5MZU9lizVmVxxcMaV9oPa0Oj1UZjbhhg72bAjv1lVYdYqZQC4Li271qz7dlCaMR+gUaU6lXq0V/hIuCcIUksPieXG53yYgKS5LJlQifJx+/vCGDn2ry24Atcg+PSfU7nknMY/qGt/nFtJnVNdcgJdrPRrI3RjnPCwOX04us5ZbCVg4uR98EFOfkKumYeuKslXJ8/fHHKvv3slFyHApgLrxuSE+23//iK/Yt/cCCsk/3x/gHCriVPq3706z/iYETX64Ymi8tuQo5H5nSzsVclvb/zm4dVintzDW62D3/OFXAFXAFXwBXYyQo4wNrJZ9/X7gq4Aq6AK3DLFMAFQyD6B+R6GZssh05/peY53dSr+6AcIvccabNH7mm5VibYOLGYXlsSnKAUihHK+gSS6NpHaLR2IRglwKSnqQ7EyWS6R04r56i/J6FyqnTIXApQRY4dnCPBkaWbeRw8imgKQeKFgroN5pptUJlRPYI+cUEa4FaTnEz3HckEAECZYpfCsrsF2QBQkAZAGMAsE2a3+gdKAg8I4rQqp+jUhULIbprXpHHItLc2y0VTFayQ+0sgbcEstuJOr0iDmlxlwKRmgTWgDYMyyXlRwuMHU8uWrKEVa8FRI/9bWH97dnlaBpj542+P2+WrrF3n8IFW61HmFyWTK5WmZVUOefxAWgH+1VBmhgusWTqeONZqBwa2zvXEWpMx6SYoElv4q45VwFliKjGlZDQtcNkm2Ai4wc12aF96y+EV8yALDcdcl/K9WpXlNqsSU3Tmsm3WHE8cbQkloWy70uA8Ut6I24oxrPM7JNcY7kPOLyWzQrUCZepsKbiVjM3Z2ctqCCDnGdc913gYOibvDYLkDyqDjQF8KinUPp6JNgoPb/gD7w3ArS5/zakOdZPxqoDh9bvUaVepZpf981/du2F4dv0e/TtXwBVwBVwBV2BnKuAAa2eed1+1K+AKuAKuwFugAJABKHP/saxyiuqlZFVZjHAmvX4+b3cdTIeAc+BS48A5BSsqlOqOlLLKxYBORd3Uc8NPnhI3/PPkWOml3MS3phP24N3t2mfWdgli8RgAAzhAJlW9HGs+AJkelf+xz7KylDr0uuPHWsLhgSCUo0WOldas9qs1FBbgQthogRhwE7/e0dMZ15qScgZpcXp9WXCBEkc6IOIUo+xwtQ59B/uV5yQIF0or5WhrFhybk35pUbLH7m+XNsunsQ8ruH5S0CMa5BrhQIvXuUn0sLorlu1f/5erdnVEdE+DdZ7Uufq1n+6XLqkV3TSUij18T5s1C3KclVMOsHZQJZSfUoA7IOVm7q1rB1/jF0DKy2NJlQimFYAvB5QOpv6O1tWWCODyyL6M7ZHjLSNAuH8grRI8ka1tGJzHjK5vAvuPH5q3M5cKOr/zAfLcLefXP/7FgTUdNXTV1HuC6yAnCAYkTCjTi7cFLigutYRgFsCKMtOY3hyU4PJ44xDrCuWzjY/d4KBqfHKVr4FnXO68LxjMky+rtYQNagKH9rUImtUU5l7RtRzeipbSa37lp/vs139mIIDe8EL/4Aq4Aq6AK+AKuAIbUsAB1oZk8xe5Aq6AK+AKuAIbUyCblitGzh1yi+LNAiayTiV009+ku3HcSPcfzQZgFEEsbt7bFkqOZtSZr6AOZuT4sJ9puaVm8jE7olDoiu6qCbAm02qX3EGffaLP3i+3Fzf1VSCU9gMQYgBPQgB4+C760BzKA6vFRQwQQa9oCz5z7CKwrOFBtttIthBzZT84WRgAglKp7uaidBB3WgQLwgbLfGhVZ8SffFeHffX7EyH/qKqsLoLyu9pSdmWkbF//4aQ9eFerQNbinzy4eaIujI27BGAlGlbGOfj69ydtbLwOr9gW99voeElh3RP2+Y/0Bgcb7relA6139cft/lpWEDETyje7O2PBKVcUqMson2yrBmD0+IGMnFYAyphdGVMemtbYLg3ukgPtkLrxAflwY3WqKyHnazsGbj4ytYBMQM9dvXJ96Zzu6kvbJ9/bGUDtWo8LZKUMlPLPOMB1FnC0AFNlaeoQnMuoJDIjF999R9L21e8mLTdHyWh9NAlkdmj9d+y53hcIQN7oQDccbDQ+AJLiPKSb5F6VvY4KSJMJ1yt931SY+/ikmjLomvs15V3dtwCEN3pcf50r4Aq4Aq6AK+AK1BVY/GvOFXEFXAFXwBVwBVyBbVdg/2DGTl8shZvxhG7MY7orTgp2kClVkQsqJ0C1T13iAsAS3QEARIPMKeBLcPAITeG4CaV72g8urGE5P3rkOHrXnVm790g2uK2i127VZ8AEnQEJhifXiLKxFt24N85zrccC/CTlLsurFCwqj0xqLeW8SsX0OMHk/AM6LJcbxHOEaD/xWKelBDKefH7KlJEe4EJHS7P96JVpO3mu2Z59ddY+9O6OoElW21FCmResIncLQBbBvbq3Z3H2lJuNCkSEwHzNEGhBSZy4hRx05XCOgBmEhC830sl6NzyO0QiNyNKyLagiBKYVtA72D5B88HhrKNPDqRQ64+kJ8bxwbABjBqfbNg7OZ4+uQ0AWULWo84ojrl/dDgFN6xk4BEu61oFC9x9ttRdenw2OOeBrv9xtD2itXBOs66gcZj//0R77g6+NqrlARccU9E3F7dETLQEIc9xmvQ6HGF09NzO4/oFqOBPZZ3Re++VipGyScsWRiWqAzmReRXB2M8f017oCroAr4Aq4Aq5AXQEHWH4luAKugCvgCrgCt1CBAZVv3XOwRe4SkQXlBXE/3d6WDIHpwKDohjgq22ucGiCHm+eocyDbtilraWyqGpxWuFK6FbyOY2u51zfuazNfAwFW6gwYOhsKoKz1+CGDSyAKiAXGKqrkjO6JrQIX0QA4lQS2KKWMBtBmNkeWF10Ra3L5JOwjj3bai6cV+K1sK7r94XSqKDQdYPT8ybzgE6BPf/pIt4tyyUyr3KsL4CInDzCos02latfxDT1GoL4O2yRrD/AqGh06Z5RjUvJYrS4P8ACNEfCKXsdnzvNmR+iAp/lHJXEE9pdvcl42e7y1vh6oNKPlAbM4twAcnF/LDeaO0265a4XzgSOvqLLZPnXS/ODDHXItqhxR1Y9H9qZDphfXGqCXa/HvfrLXPvSudvvT701onzVljWXtvSfawzUDbCJLbSvLNpdmn4WMO5Wtsv7dfVtAJ5cTzB9zBVwBV8AVcAV2uALL/0Wxw0Xx5bsCroAr4Aq4AtupwAk5pOZFn149OxscWACbhMrlunQjvpzTKJoL2xVVYmfNiyQF581ehYLXwZccXYuMJ3rZLftM2DnQKDjENE2Cw+nWd7OBe6u5OSEXF7leAk6K30ou89cJzxXL6qgnmMVa+Tcr6IWLCkhEeDhuKJxVAC3ABpArADVtMKLg/MvDMT2XsjMX8tK+EDKWcOucuCNt9x9vCy64WnlxvgDDD7yrzU6+UZD7p+64osKwLZu0D7+n/Zq7ZvFs1FcKMJuWFsx5akYZTjoGDjvmzNeAr82OkHnGAhsGoeWAQKARAftou1nHUcPu1/Qla6xfwys7rjgnk3LOkXEFvGKurQKxS8tQmTuOObr7oVtne1NweOEAZNS7aC6er0P7UvaPvzB4wzyXA2Q3bOQPuAKugCvgCrgCrsDbXoFl/kR828/ZJ+gKuAKugCvgCtz2Ctx/LKMuds128arAiBwjnepQuE/5QTcDWLg+OlSuRvc1AA3fh/I93dy/lQNgNC4XGEHqgLh6vlOTHDNymMkxttStsnSuAJ0I6pANViSMvmFQNjkxrXwjARvK+QA05GPRRRDwQUkf5WFjk1UTc4Ix6QPd4aSRtKGcjnwo3FpvXs7by6fzllerOBw5PPf6hbIdOVgTQFHWksK3owH42K38pt/4uQH72g8mbGikqi6JCfvwI+020FsPQgeiLIVEUwsZSTWdI7KwKPMjc6tXpW+E0wN5NjsqWsvr5+jgWLRJuflaWpO2ty+m7pBAUP1TzhoNAnAnbTXAwdEUBfnjtCJ362Zruha8LiAF3Luq88a1weB8ZjN1EEcpZ1SKyvXAdcVnnHdRMwFKFH24Aq6AK+AKuAKuwM5UwAHWzjzvvmpXwBVwBVyBt4ECB3enlGOVCqAByLCWEie26xCc2O4BMIjytsijWgmCRKVsuKGAalUFY1Xn6plVGITYx3oCy3HhjCkQG+gB7NA0AsgAgpH3VdBxcJl1CPiBPcilAlL1dSfkwqKkLmNFsZHyXMma1JWQLnBtKvfbJZdaQn/1nB+qWkHwCnBl2g/uH/K3XjtbtAeP3Vj6BSw7sDutLnJy9ojS4KhinQzg1VLgiOsKmHfuSkEZXLN2WdAL/Q7sSdjPfKDHsspF2uxAl5dO5e3bz07aS2fyIUC8pjUTav7BhzvtbjnK9ilrbQBNpFfnJq8XYNWI4OC8Gg3MCwzihGMgISAxlRScU0dJYBPwkewwXHjoMzxesTPniwFaAWyP7VcrS4HFCHihfXCN6TyVKoulmKH0UOV4OM2atD1uOEoSAWY+XAFXwBVwBVwBV2BnKuAAa5vO+5z+gM/lctu09+3fbU1ugOnp6e0/0Nv0CL5+P/879foPmTT6+bVT1897n5/fO3H9gBfK8MYUBF9rUo5QcDvVf0gD1sikSsmhtHRMzVL6V1vIo6qDHbaZA0YILNWUD1UR2FjrmFJpWUUQZEb5VgAp5gW8OjtUUmA7HRiVfy6YgfNsUC6oOBSKoakd2mW2p7fZDu/J2huX4nZltGIJzWG3nEmx5opcWOraOFsUMBGECtlLel0Np5bcPqWilYpxBelXbnr+QYdAF+BcQroU8maLPQot5JH9+NWCPftawc5ekZ1LG6LD5KzC6vMV+/yH20NOGVPe6LgyWlWm14w98+qMgBrgkFJJQTtp9c2nJwR9sjY8mrfjh1IBXs0qF20paOPYgeE1nNJyuXzD+s8PVQT3SgGSFZRHhdust6MOKAF5QKou7X9ysn5OojVNaF5Pv1ywF04XbFwOMcVDWYvccj94IS7I1mqE7F8HRZVVNj+ncPRKw4S0s/CHKqdYeuel9XaOUqmkEtXyTc//dh7/rd53sVjc0esvFAo3XP9v9Tm5lcdn/Tv19x86s37GTvz9z7p5/zfrvxDwbycOfv7zN+DtOmKxmLW0KINhBwwHWNt0krkJ5I1wu47bff6b1d3Xf3tfv37+N64A1/5Ovv7544V/t/PP742cfQDIxDT5UcqRmpmz0ZlccC71dcvxssATKsqG6my7sfwNdw6ZRoAawFM0irJJAbzmBa9K5HatYTCPnHKcGHMiTMmYHFiCJJeGNadxZV0JLtbkqioLjtXm9Yf2WM0O7VGJonYPWOEowJVEj6BVj2CUgEiJkkDNLa+srJyC3Xs7m+TCooOcwJLAVbOOwesPDKrkrip4t47zr13eMIZGKgI2ZZUb6o9hKJmGKuE0v3k9VlCelroF3rVxFxYZY6NyQ10eKQqeyRGlfWt5AfDwuagulhevFi2TqNlgj5xqWbmwVK5XkysucsJRpjmi7oqUIZJRNqjzTKljVS0LG9c/rdD0k3KmXVCZKy6pXEHZXnoMEMVrsuk6NBsZjwliySGVjktfnUOd75PnSnbyfMmGBNuYZBWR9Zlr5LnXZu3hu9Iqp+SM1cecugbO1+bWfK1Er9vKz8DLxvVv5b5vh32xfq6BnfbzLzo3vv6dff1z7TN26vXP+oFXO3X9wFvG7QrwEnQ42SHDAdY2neh4PG59fX3btPft3y3/9eF2nv9mFfL1+/nfqdc/8Gp2dnbHvv/5A4b/CrvTzv/QWMXmmgRdJgr23KuzMiUJHgk+Hd6Tsvc/1HENYhEyT54TZWF05turcrikwtfJKmJQ7lXU8ziU6JZI/hLlc2sdlBvGU/WbiOa4YIr2gwsrNl2yrs6ySszUhU5ELS7okpClp01dF7s608GlldBxKJmLsrSWHnNUzrKrmvejbQoEj8/a6YulAExaBXjuPpK1Dz7SaZkkDrzqps7/xbG89XQWLJaU60hAhgEETCVjltC/ZEu79t++dHoBQkVldTc82fAADrWBubK6I0rzZpVJytlEh0TOB4PvY00p/ZdYZax1ttkulRIyOA+UnhK0f25EHRnVCIBSTNQeyzfb3YNZATfljAngtbX3hPNcUClnRaAQWDin/5Xn9f5Q+eXVyZpcWbiq6lCqSWWE/Z0prXve7j3SEgLri3JSlUUkm2I1rX8BhGqBCdVxVmop6+/vFqAM3C1kXO3qS94Q4h4mfgs/5GXxmpqa2tT5v4XT3fJD8bOf6oGd9vMvEpK//bh536nr59oHYvT0iHzvwDExMaGfw/rd1d29A1ev3wNjYyrHj1lnp/4rxA4cIyMjlkoph7T9xt/PO1COt/WSHWC9rU+PT84VcAVcAVfAFdheBSjRoxscJV8nlVU0qc95OTFEMuzKcDV0i3vPPVmV7MUFgKp26rzKbLBKabykMPSH7srKvRPX5vMhSB2XT1qd49rVdW+lwU0CY2nmFy6qZkEOwtpj+rqmuTG6BcKKcodlBF3IRorpvxITuE5we4tykaJBHtZyg8PxD7gFtPnQIx32wLFqKG3rV7D64X3pkKlF+Vg0KKeck9sro2NwzLWOlowyt1QeR1kdIeqicAJNZG2xr6QdUBZX4wDQkSnF/IBMhNKvFnqf1ZofuqtF5XgzVs7VYSJOLP23Mx0jZgM9CcG8pPV3N/6ZV9cGiMc5Z1ACyH905nyO6Nz26O/2aZVuFmp0XKzZ0Fg5NBkAmuWkx2V9PyzYmVMpH8taHHLIKZi9PRtXCP68wGdGa6jnXHE+G9VL6T8Sdyhkvlv6ZFQKGheIbJdePlwBV8AVcAVcAVfAFVhNgca/bFbb1p93BVwBV8AVcAVcgXeYArieCEEfnyrZ9GxZWUdy2qiUblhAgudeVae70wJbRw6kVI5XL1nDTQQswc3zgxdNmUbtAi+xALGAMEu78kWSAblm9JpK3WQVwFCbXhc5j/hMeRr7pVvfjEALeVeDKgksVxJ6HV3pFPYtzsScgTSNIwpXb3yMr9kv3RFTCYLp6+ilW6HjvXJsAcc65RaLBnN76qVZuzhchOGF0ro7D2bt4K5k2Ee03UqfcYHRcfDBY232PQGmvMoW0aOzPWWP39+mcPXFObNOHG3RYG7TyvnqaphP9Fz0GT3Y7t7DLfYLH++2P/zGpHK9FFmgLC/A4fGDabvrSMb27RKUE1CKBp0CGZxTRrGkUnFqCBfGyERVZaPzdkHlmrXmvF0ZUx6aHrs0WtT+BbXkrhufmNP5i15x/eec4mMSMYFQbXM1Xbb9g2nbP5AW7NJrWaMIXVJz75Br7sG7W21P3/Ug7/q9+XeugCvgCrgCroAr4ArcqMDiXzY3PuePuAKugCvgCrgCrsA7XAHyWnEy4RCamgZUKHNqXOV6CzGOBTmfvvn0lL36Ztz2qGSwp4OsKL1GIeo9gkDNzVUFiSsgPamytQY31HKyzQhmRACF5+lWh2uqEdjQhZDOdmXBqo62WHAmTarzX4vgzMRsJYCUtmwdXvF84yBQfKWBs4nOivPzcgkJMIFxOtXJsEMAK3I8EeV1+qJcZyoj5fmKYM+08qKmZ2f0Xav1dSbl1Kq7hXBMLXc89ndQXQt7OhJ24mirnbmck75xO3qgxfb2xoPO0RxZ/9IBhMMhtRIExHk2Nwf0m7MPP9JlDx9vV6B7Ts6nqoAe3Qfrc0wJJAL5cEBxriIXWYvO9djk9fCq7pGat2dO5u31Nyo2W8kJbpF5pbXr3E4IquHSo6PhzUZR867JChbXedjTF5ezKm5tKlsEgJq6F+7fnbT3nGi3d9/VerPd+HOugCvgCrgCroAr4Aosq4ADrGVl8QddAVfAFXAFXIGdoQAh3pOCJvvl2FHRl03LYRPBq0YF6HxXlAPr8jDup7iyIpoEdOJ2ZG9a+UtNoQyOzoArDcDV9Gw1bJsULAMm5QREcAERTE6gd2sL3QIFXbQbQFY0AGMAlcFKMrip6ERI90P2EQ3QFSCuceCgorQtJwcQz7QLPvXLIaWkrQDJIudX9BqAzZCcR/G09qv/k7nFFxWV043KjQTwuqoweUoYQzC6oEyXgB55X42D7C+6/u1SFOa9R+sZVI3PR1+TFbaRgc6sleUT1H780OIxAGtL19V4DLKmRpQHRqmgcJOC7KW3QNfQWMlee7MgB1bFilXU1PkRzMwruJ0KSo7VYNhq3OW1ryuCnjHtb7fcVUcPZAThYnK9tWquKhfUP2gkoNSHK+AKuAKugCvgCrgCG1Hg+r+4NrIHf40r4Aq4Aq6AK+AK3LYKkGs1IshDAPu+XSl7Vm6elcakHFrVFpWqKaA81ixH1LReq+6FM+py19bgvqLEraASNQATQd8AKWBVTs4nBo4fnEZ5gaWCHpuSw4osJIBYT2ciwJ+lc8BRRH5SNDrbmsLryXDCeEVWFccCWuHewnl0abgc5ibeFsBYcbxmfV1N1t1x458/lPPRiRH3U0VazKskr0k5ToGR6TEG3fvIg6qXQdZCeePYVJPt6k1dVxrItrizrsdpPHr9oNySsszGweskxaoDSKXl3jB4/GaD5ym9JGy/pM6SuKVolnjuStnKAonkklX1AOdPpq5QrkhuWFxdG1cdmnd3V9zuOpwJDjTKSkPQ+7UXrjK5a9v5F66AK+AKuAKugCvgCtyowI1/wd24jT/iCrgCroAr4Aq4Au8wBQAYZ6+U7PyVYnDgUJp2WCVelNrx3HIDhBFcT+pKh1sLt82cAsBPXyhZayahnCdK9OpB4AS1A8Uo18vJxdMjaBS21+O4riYFiwhqx9VFDtacXE5j6noH9JgVMGltAGLLzQXQs3QbHFPANCDWyEQ5ACwCz4E0aWVqpRW7NC7othRgAZHIaaLMrqu92SbydYhTDCBM4erK/FIlpT37yqxK9QSuBNLa5caqCmw1yWk1qdLG1hm9VvlO6xmtclKRCxaFqrMmAs2XhtuvZ5+rbTs+WbWzQwVlXFW0Dh1L56ugEHmAHzlhGWkA2JrVvMjq55wBsIKbSl0ayR9baRxUx8Nf+Gi/PXg8u2x55Uqv88ddAVfAFXAFXAFXwBVYiwIOsNaikm/jCrgCroAr4Aq8gxQAUJErNTJZlsNILiOBmLzKxjrVGW5vb5PK5ZZfLE6qpP5yaBfkAQoRqI6TCsdOsaS8JDmUgB3AKxxXM+pcR0NDutzNyNiVEUyKCS7paUETlQ2yP0EUQBdd9IBMBLXHKnLqLFbFLT+ZJY+yz1kdn884u8paE2sDTAGKCppfXMnzmvYNIyqNY+57++PWWkgFuDen11LyRjfCP/zaqDryFQTY4jagfKcj6lx4QEHluJYsSyi6dtt2w65v+gDAinLDED6v3Wx3eR3n/RtPT9ZLOTUznboA0NqUVUXJZlOH4GVBIiQURF9Sd8MmQCYOuqYA7arVZpsUHcSZtXQM9sTsZz/YZfcecXi1VBv/3hVwBVwBV8AVcAW2RgEHWFujo+/FFXAFXAFXwBW4bRQA6lAaN7/YhC7kPE0LOO3uaxa8MbmKGp5cWFlWDiZK/QhxT6blzBF86lewuXZlZy4WlTU1p0wm5VTp+4vDpRD8PSfgUVNWFO6mpEAQOVKU+xEWPyfAdHWsHLZrEjchN6pL4ef9icUaOhxKOQEuMrTYL9lPgC7GJR3jhVP5AMh6u1IqjauXH1ZVwkjOVrN2WoPSLAwC41MKgF86gFqhA6ByoejOd1i5Xj0qhZuWBi9q/1/6+mjojFh/XdXOXRWA0/wJlt83UO/IuFLo+tJjLff99WV2y22x+cdwzv3NM9P2xnmFnEmXTDiXKpPU10JXoRPjJO60TpVEprPW00oWlgCgHHQyx0mLiq4RrbllPuSPoY1Yo8BWk+1VrtZjD7TZx97bKRB2o76bn73vwRVwBVwBV8AVcAVcAf0d6iK4Aq6AK+AKuAKuwM5SAHhFhzpAUEElcTimZvLzNqayu4og071HMurGV7Kx6UqAXPOCHB3KvurIxK3WLLClfCjAEPAiLaBECVxBXfFm88rFSlbs8tWqDY2XwjYEvEOeUgIbhLFz7JTcPgMCXa+fy1/Lk+IMAHJmBdG65QSLBplTwaGkB8ZU/sZ8JzSvv1ZnxHOXisqfSqjsLyEgU7a9uxL26D0dISie9XVpP+RaRSMui9VAz41lfqqSCx32cJLh4CJDa2Jqzs4NleyrT040wKv6nnBbnTxbskMCXXcfjoXyuqg7YXSst9Nnzu/pC0UF8Mtxp4nVBBVnCqbzoU6FyvmiJPLgnpRdUkdJeeTsjv0ZbZMOoe1XFV4/NlGSBnSdTNTBo05iUbY1yg4H+9L2sLoKvv9d7eEcr7RuQGTUgRLXHW6+WzkAeFG3yVt5XD+WK+AKuAKugCvgCmydAg6wtk5L35Mr4Aq4Aq6AK7CiAsCR05fKllTo0KF9GTlfVtx025/AnQRQIJcKOAQkIvAcV834NEHec9ar7nqd6tpXnsOhUw0lZl1tSbmmagH2UO4GkCoSwq61UIK2W133ZnLzNizQVJVjqkBtnoBQSzou+DRne1SeSNkhHenwU3WofG5KmVWzMgUBNHBM0amwOqc2gxq4oghMvzJSspPnywFuPfPqtL18uqj512U6eU4gSdldD9yZVZ7XvB3cVbLdClWnFLFNji5K9BiUwe0SNANQUe44NFYPnw+dAKVFQq6vqsooOS8EuBdUAvnGxUKAZvUjXf9xVh0L6UbYJw1396cEsd7CE3r91G74Li+3GGCQDLC4oOOczo1Ou5X0OI6rPunSJzjVls7Y5GTR7j7WoutBwf7SqaMtZuO6FkbGKzajc43rDDcdq8VtNSit9w6kbgqvAJxcW5R0MjSDkCHGfrZ74Db8zo+n5dTLBWj7+Il2e8+9rdt9WN+/K+AKuAKugCvgCmyDAg6wtkFU36Ur4Aq4Aq6AK9CowA9ezNlXvjkq1wu5QgIpcq384ke7rLcrKdcR5VwEdze+Ynu/xnkDuCKInLwo8qom5Tgi1H0mV7GsyuzScsl0tiZsz0BSMKpmirJSYDsldpqvwryBD2RXTShLK6WQ8972uELMqwI5ilCKCU/p9VVcL1qXuFTIzGpvVYmg3FKM6dmaZTNNoXshwe2QJSBTOhUPZXywjhG5f06dLwhMAdfKduZS6f9n772DLM3u87zT3Td3zpN68u5sDgAWiQQBkCBFUqJIipIsWkWVgs2Syi67SlVSla2y/7H+cNkuV7mssksuWZKDbNmiKFo2RSqAIAmBwCLsYnOYnDvnvn1z+3nOnTvbO9OzGGAn9Mycs+h07/2+75z3fI3t++z7e3/hjfcr8fjtn87yvHBq/56uCKh0lm1tAck4h7BMZ5cOHLsgGtj+yruEsbPmLWCcDrEC899HGVwx154DEVBhJeZ3tTO1tl/r+ves6Vmcau7hboZXzlcQKMTS+eS+rrOPZpkJISdH8+HTT/dFuJOnPLS7hStLpxrOtrlFAt0tx+QYg+/NArsyVwmrZULvAYt5wOTeMUspP9pOJYTswCvnI8byMe9Dc7n8nfB+8vfgTgzvbe9Nyx//8b+cowx0nZuwDd1OAUJ1lf3JLwzdiUulcyQFkgJJgaRAUiApcA8VSADrHoqdLpUUSAokBZICj54C88v18Ftfm+dNf7t92xYg5dyVjfCPfzeEv/ALY8AfuvIBVYZwutzN7nPblRcUDQOcNnAR6UCanq+FczPlsLhWD1XK46r1RuijG56gobwJdCLnqtVsAjR0S+GqAgZYSbi4APDiZ4PNa8ACXU/DgI4CbqYtAsB1SZmL5Rp1RcFHIkji8sAKQRew5JobrAMYhCHCjLmlepjB9eOH4MmAeF1BOw4meu7qZpjamyOTKh+D5C1TdHgtP2oRajTDm6fK4QKlgUwVp1UL+ARoa7Svs9VqhI2Naji2uhkdR73ANDsdusYbxyGuc2hv/p6Xwt04j9v5WYBlSaUAcf9WIaywzxUem5oohJ/5/FB0VcXz4DzrDDsUTo5eg4k8aDZYgzD+WiOLS69MV8ZWKLDHl9iT/t56OEIJ4k5DcGj22I2jzL3nvNr5X5YXBs6/FV1zN75WIKVj0Nd6737U0O21gotPeHn+SiW8/l4Zv5flqW1wVqvXwte+vRh+4ceHvOXSSAokBZICSYGkQFLgAVIgAawHaLPSVJMCSYGkQFLgwVPAEPD18gfgxTfjvrm+Ol+JLhFBi49V6GZned29GjEcHSfNO2cq0dm0udl2rcBygD24k5iK4Mlg9SHaBc7jWqkCMXTMWAS2wWsWyJcyiD1sNqLDRihiRtWxA3kcXbi6AAlrlAh2U2pXpWxtoL8bYGQOVlsDIcnYYI6SxAaQw0wmrkVu1cpaE6dPNbq7hBx+NIArNc59qyGQ2jtRjC4eIYYfEXoAyFbJ97KEcw0ItkRQuV/VXAfSBteucN4GoKwv7/qa4eJMNQIPVAhPHSmG7+LYcn6d0U9p5b//p8bJ6sruaveVGljquI7brKQrDXeVbjEzxnKAu0/hvLIk8FZDuKejzjHQG8LsIqCJ+7SPY4YpzRSKGdJ//moFh1ZPGMTRd+OQNwmKvE22D7tQDt5QQug+lGIu1we/B2evVMO/pQRQKDVAt8RPPFEio43J3GLEjpbXrnWVMtGG7S0Z3gvCL+fh/XR1oQ5UvTkP7RanTQ8nBZICSYGkQFIgKbALFLj5L41dMKk0haRAUiApkBRICjw0Clg/t8O48eGdXCo7HHZHHvLNvM6wM5TeXZ2vhlOEqS+s6Mrh9IAaKgAjsMFwE51LZiRlAW2rlKGtAkSkAKsALx05W9CJdodAg93pTmhpYTUHfKpHcJIlJ6mPUkFf+503y5TpdZOblIuB2pYOdgNBhB+6awQmZiVVvDDDLoktIEyG3LAaj40O9xAOTwvEHcZTR3txFGVxbDUifNIN5rBwTKiRYx7CNqFUjXMabN4AolR4rgGMK1BTZ9C8+VeWzOUyOIeEMv1ZSiq7wyVAShXdHp8qha98diAcxIHVB2xpO4h2mNB9fkhIs0pXRdemFgXgaJbaSPPPxEN9gMli/tbwaqfpZ9FFx14nuL/M+VsAIuHrDEBoJ4Clq9DrlHHRdYYQyUaTO7mpvDc7YeuXZurhN+gAuVFp7/niSi3eVzkg6OOHC53Tfeirx3fGXhxkBs13gGl8nMXnc1lKThO86uiUviYFkgJJgaRAUuBBUSABrAdlp9I8kwJJgaRAUuCBVODFE73h37wM0FmlNm/bOLS3SOndBwBBN8u9GgKdZVxOgqhLuI2mF2rRIdWEE/j2X9NKC/BhlpWOGCGIeVSxfBD8IYAwD6thXhJd7NqQwJ+BIyyjjpPGkrM6risD3GVJlhOurNXInyqHg3saYYzQ8FEAkQ6gfgCRzMp5GTReIGdrAddaB5aUa4AiiN+hvYV43tOE4ceJMlev9+SRQnj6aJFSyCr5RvVwiNwuy+UcniNmcXH+Hs6h66vER5mAeQ1AMBjW0B16mYNuo25gWzaLywi3Ty+lkf2c57GDTIoue6PDbQA0wVfD4c3a2q2j7URqw6AMzqOoAdPV5aYLyw+D83+YoUPQfKx6A9ca+2u3Rt1s5mmdvlRmH8nEGr8ZDJl15T5ZKusQcup+uxHa8hKAYHtS3mPffnPlOrzyONnUGplrr51avyXAiiWh1+xygtdnjvcR4L4WYZlzyABJv/xib5yP50wjKZAUSAokBZICSYEHR4EEsB6cvUozTQokBZICSYEHUIEhyuZ+9WfHwj/96gK5SjUAiqVu+fArPzUSIYJL0sVjR7d7Nao4ZnSlrFFCuLDSiCHrXlt45RBgZaE5Q+RZnSU4/dTFSgxyb0GhypTjaeEpEnzeBGQ0trpjxlB3BEFduFsoAcRFtVG15FC/E+HhuK5yPZn4s86gSo3OhetAFcr49o3T5RCSUqd0zxKydeYkzJJBOB9Bi2WMQibLC7/y0lB45/xm+M5b63TRaxE2ngsTI5kI14xwKqGjWV5FsrQ81lECvAibBHQjg9lYTmaouWOA4PkI2eIFCbSnI6IOIAGV17WT4RJdFV1JF5oNEGzvuf3YzQPmeH3otlq3bBICZCZVCYjjum53eC6VtNOkzrM5gKelg+6znqwCTSN1sF0BIOb5Xo1vHCX08qMzckDCFWBU557TqaWjTcjk2OR+KAPJbhwN9qBOt8hbDa8hWBOOmaf2p35yJBzeVwynLmxEaPb55wfD555LXQhvpV96PCmQFEgKJAWSArtZgQSwdvPupLklBZICSYGkwEOhwDPHSuHEoSJulXYWUY43+ZvAHWHC/ehC2HHerJIr5Pe6rCo4kiLDgRnYhU/AsL4GsAI26bDRGCNO2MJxJbHKADNKuHk2I2QwjJvAc6CIpXlNArQKALB6vR7Bl6CqgRtrGLBh6ZjATkhlaRsMDJCmu6uBG6xBNlE1BnqXMD3pSmtxLjsJlqgY87gcIGoKh1Wmuz9cAaToDlslZH2N0sYCeVtDlPz1YuPi0gAsPwBfQEQL57ooRZzaysf8Jh1Kuq5EIbPkX7m+K7PVUIGgCGB0iB0lmNzXWWq4zLl1kuk4GmSd14xCu/b+VKualiWGQG4QICeIGqZZQCfg/gdN3n2zpNP71GFp5f6JHPdJX3jzTDlqbDlh243WhkrLdJccGfxBZ26DSQP/266sdtmmQeudkeHGtDPkqYsQxW1Dt9++sVv/+er9bD6X53Xezu3wvgL35fAP7Tjbdtn0bVIgKZAUSAokBZICu0CBW/8FsAsml6aQFEgKJAWSAkmBh0UB3UBmPXWGpVz3a/TTYfDqPMAB2NTOe+ohwJwSrx47wwmKcNSQW2SPwm6sVVtAJOGHHGMLqCEUGsSJZCe5nm7dSZTWcQwPx8cEHx7f3ZMF+NSjG6YbaOc1LB1bBYyNkWclbOjhuAUcTn/wymq4QjljHWePmUl2MjT0XWeOAeSCryrnXSaEXfeU0G0cyHRhuhIWlmrAmq4wDhCxbHEFiNIHWDkw2S4VM2tphK6LQg3PZ0lbnjK2TgbTY8DFDZxfx/ZlwywdGffv7QXidZETZvYSGuHgmtA6d20YSB8B2H3cw85cbvXVNQoI2eLrQ4fT7cIrD7LzYwde+bPuuTV40j4glnt8GQAoZETeOHICR2tFb3MI1kq3CJE3s+z4VIH7tEh5Yo3z4vZi76b2FMJnn+3/yCvwspvKO30sjaRAUiApkBRICiQFHmwFEsB6sPcvzT4pkBRICiQF7qMCQgzf1AtFDMk2X2m3DMsEzRra6Y37JgBIh5WQaM9oHqBEGR15XDl4TZXSrdGhPGV5QApLBglrN0Dd9/+GqYt0LEWLkAsXVnnT0HZAVsmw9DbcMrC7iGNqbKg7FDLZCL/20PHNIHghWBn3mXAkh7VrmrLKl19bjWWKzlXA1aAVYiVjQHgWWEIQt64vnrQLnXCrCIiBmjEvugVCzTCPAWu2wiJdERdaNZ7PhG++AYxj/p96cgDY1hN1sIzQD0sV1cch0LF8c4D1l0uUEG4EQuS3wvnZSpgnmHyD6LJRShf34gZyjx0GhXu9ezns/uj9ZgZVLzldOqw+avi8XRIt11Rzuw56P9zucH03ZlR5rPe7pZUThKAv0AjAYHzPKrBsB/ZDGu/Q2I/mX/7kUHhsqhHmlqvB7LEXTpQiQL1Dl0inSQokBZICSYGkQFLgAVIgAawHaLPSVJMCSYGkQFJgdyhwkkyor35rBWBSj+VmX35pEFhgGPnNzo97PeM5XEPfe3stzBNmLlo4QGbRj7/Qz5v+NrxYo9zuylwtLFMaVgI6ffKpIvAhhEvT9Qg6urYq4fCBUqgBbnRHCXvOXxUmGdbO+ghlt+zREkC7w1UBHXV+WF0DgBVw1HBVywIzGTOtyIoCGI3hjNJpI+DTFdSHA0z3ld0Ll8mrujwHsPKEjDypSsKXJsAml+H8nEOXWBvYbMVQdcsULeVzRQOswWs0qH9coyTSrniWCy6Vu8O5y/VwYLwGkGNjtg2znMxt2j42NltRl3NXGmF+3b1tQ5M84EcNCcCKsK/IGtWy0ylv+znu5PcCJB1jOqjWAVeWezaxwZlbZgD/xAgh+NxzEwDIW+WnqfGtHE4fd679AL8TR0o4pGrMsR3kb7nppTOV8MapzXDicJ6w/p07Bd7utXV36fbyg5Xc7mHpdUmBpEBSICmQFEgKPKQKJID1kG5sWlZSICmQFEgK3B0FTl6ohH/0O/MEULe7Cs4tVcMlspP+0i9OxJK0+9mZzlyt77y1EbvxCT8s81qm85/Ona98ejBCI91WhlzrIhIQGMo+Rai8bizL5YpAo4nxXmBUT6jSCXBuBXcV7qdVzmHnQTsnFqgXnMc5teoRnKcBbGl0NcLmmiWElKkBlPq6MhGIWTZpl0Hh1aF9ebKutnieMHTOo2NqE7eWmVuxRJH5NrBT9ZeyzKQNq/pxXQlx9DtxebojMkfKA3X/6CiKriqAVBVnkEDMboV1XtOMIeDtTK3D+wuxbPBWd8Q6pZBLhNlPLzbCG6eb4d2LK8AsgR5rR5aRs/nwx35siGylLFlSlEtyXaHZ3RoGmC8sNyPsqyPMhelaWGLf/tFvz4cZcr8Ed3193eHP/7HJ8NSxRnj2eCnqcCfnEyGdIBEtt4/tLkOdbb1Y7bzX3jtbDV/97hr3lR69EF59P4SfoHPjZ5+9jUCs7RdI3ycFkgJJgaRAUiApkBS4hQIJYN1CmPRwUiApkBRICiQFdlLgG6+tAYQIito2NgA9L7+5Fn7600PbHr33387QBW5phRwpYI7wqjOm5ypAtnwMJu/BPeQ/jhVcPb5qHfeRXelyPTh+MEIdPVAgQDsbnT5XOadOnwrntMObeVinL9GVkAOFYJ4A4w2lh/zM9z1dLVxanI98qCP72q6rDPWGlh2uk02VAf70mnskFMMF5jmqyNkEPlkKqGuoi5LBEpBsalznU1csg+vC/tXV1YyljuZZ6S6rNys8Ru5Ssyu8e74MTINw4ZTytU1Alq4toZaAxUyoW41FIJ3OtDdOlsOrp2rh8izpX5yqm2laKrlEHeHLb62Fzz/XH0PBLaHz3Lq2dJQ5hGrb4c6trnU7j18FngkZHTOUMZ5ibf/nv1oEDrYf8/GllRYgdS78+q+MxyywwwTO3+nhOnW6eS+ps+BOgLl9qEMP+/lHbyyzl20Xnc+r37de3wgvnuj/oXK3tp87fZ8USAokBZICSYGkQFJguwIJYG1XI32fFEgKJAWSAkmBH6DAynptx1esAWM6oeA7vuAePGjxnBDpGvu4fsX4GGxBmGPpWy9ld8vrQCd+XsHZI6QYwE3TEm4BKpqAIDOXjgBFJilVW6Cc7p0zm56dc3dFoLVCyeDqOlCoWwgFsOC6Dr9aQjhLsLod4PrIo1qjvG+ec5y6VOX6ITz/WCnmGRmUbhZWwVwnYIm5VPAQjsvxGl1TWbKPoFtCLR4HlQBDurgG3eWYbwPAtUQJYpFsqjWcXFU6IsayOa6p68uuiiXO2ymfbM/ww5+dr2t9+8wGpW/lMLMosLn2GtYFG8N5RrkkwfOW4wnPHCvst3lQnVHn5/5eSixxg32cYTB+B14ZoF5hj155B2i6DV51zr9GZ8Tvv7cZHjvIhe/CEE4Ns17vH5Gne7DTEIRuMO8bRxUiKYA7uPfOw7Ubr5V+TgokBZICSYGkQFLg4VcgAayHf4/TCpMCSYGkQFLgDiqwl8yhq2RICauiA+nauScJKR8g2+l+jslRg8a7gUIfnsUgIMmSNwPLhVhCpb3jXWEOuCAgMQjdIRiyQ6A5S1qx+kvAJaCNHQbHRjK4pHBogTJ0OnXjxBLaCb9ARW3IxGEYn+IQAtVhT0u+BsBRrUOBeE4H19e+Uyd7qy/CoMUlsrjgL329OSBNiGHpR6ZyALVcqGHtMsjb8kBhjnrPLJI7xnrM5MoTjWRQ+SCwyu6Jo3RGnKGkU6DmWoVigzivDJXXgbUTYPS8a+VGeO/cZgRZbGuENfI4XUQws9DFRxZI1gc0c1gyuR1exQf5ZKmmAGuTrzVeY/ldCSfarTKqOsdt/yos+uDewmUGBNqs7UyOLLvcqJhztfPz28/7cb5Xk48auvcy1FqaQ7Z9dGNhs8wwjaRAUiApkBRICiQFkgJ3QoEEsO6EiukcSYGkQFIgKfDIKPCVzw+Hs1co1VutXcttCuEQYdVf+sTgfe+O1ksZ22efGwhfe3k5rGxAdEAxAp6nj5Zi+Z8QJ5ehlA+3TAub1jyB75b1WbKXy2bCylIjLC/jsnprPbptTl7M0SkQOBGfd4u7gEUZHiPjqp8SwQP56PbaILMp4BDSqaRLRxdUN6V8Ork26WJYAV4JlVrAMjsKloEu3317IxynVDELhOputUvy9ox1E7hOt0JgSIHMrDIOo/kVOt0BgvqAJDkcYpbX9bMO87scwqHuQjYc3tsNcOsJY3THqzCfAeY5yBxHBomFh8BY7idouXHoMtLdtIl7y3/y5HFVMdltLyEs5TPh+MFiGO5r/9lk572dhoBNoGc54iwZYbrH1jdaQLk8OVXF6H4TUOkI28kVpqvr0kw1BuxbrqeLzHJIyznNXhNYbR+WN544kMPh9PHC0ref80f5Xlffc7jqXnmXELRt47GDBfYg/am5TZL0bVIgKZAUSAokBZICH0OB9FfFxxAvHZoUSAokBZICj54CIzhK/uM/ty/84asrlMU1w96RnvDS031kA93Zf6Xq8FkBLlVxIRXy7dI1wdAPGlN0bPvzPz8e3qdToo6pQUrALH3rE/oAGhqcw7K7r7+6GpYBRFipyLpqhEVgix32NgBOXZl6BCwz79YilJsACh2inLAAeJkBzFj65zmPTpXi8+t0I7xIF0OK7eL0hFjwsHiOBiRIGKTzS8BTARRVgWe+Zn6lBpTqDkNkbBkM7zodljFaFrhJdpXldN38bIlaF26mEtVoNabdxxwyXKNG2aHGn2HC4i2N1AVlmP0gsEnXWcedpCNqpxHnhMspB7jq5ZoDNLvrAmptEDYvIxvA1fUE8Oqv/MnxCOY8h9DLOepY2z50py0QBn/+ajW8e3YjfOuNtTBNKeUWog8PZcJf/sUxYGJf1MXMrD7W0xnThLOfvyIUbUagtryqXtlwcLIQnjhcp9tfPZw8T/bYNYhlKeZzJ0rhz/7sGNCtDfM657ofX3/6s0PRgfjuOZsbdFF+mg1fIMQ9jaRAUiApkBRICiQFkgJ3SoE7+9f2nZpVOk9SICmQFEgKJAV2sQIFQM/PfO7uBbYLey7jNBJeOXRMmXd0CKfNjRBLhiIM2j50Jz15uAhgASkBjQQujgpAaJqywT94ZY2gcjrbkUtV4Ro9OI/qDfOzstHtZKi55WBlSgkzQKtVSuwEakf2FSlZ4zFK2oZxN9ntb3I4E5451kvZ3DquLzxM2ISKOJYESuZVYY4Ky7QrdB511mXulrMpCYr4RneV0GliqCdmZlmCtx8I5zp1XAmvfJ1rEUzZCTCW5gGndCdVoVmWE/bRRdGOiDqbanUAG89vD27npTsOrz1Ip8P9owVKA6FDW/Uw3MxxvW6cT/nwk58aCJ96RkD5AWwSBArGhH2d4fwMzzdT7OJ0Nfzb76+FK7Md4ASInGuEv/sbc+Fv/ZVcLJO0zFCXmfoKwi7SabAF6ROM9QK2spm2K+zJI+SQUba6hxLVbwPE3qLUscD1f+mnRsMf//FhOjp+MK/OXO7X1888OxA+8+z9unq6blIgKZAUSAokBZICD7sCCWA97Duc1pcUSAokBZICD5wChmILr8xashytCNRx6JwaHYQIMeo8ZwD6Kq81b0q30RBQSWgjBHH4pQOv/Flo9dbpcjiLO+vidCV2IbQ7YIbgqm4gl3DKjn2Nlt39gFHkVenuyeOOmlvmIl2V6OQSMq2V4xVCC5hycDIbupq9YYG51LRHSZ9gQb240oRAZlitEhpvSWGW6eepfRsk88q8LlemW2oUh9JxShJLBLfn+OvkAhDIbC4hVidUfZGSR91iur8stxvs6w4LzEsI18+1RgbbH6X81k3lguqz01BHg8oPUYqXzTXD5Ww19JPGfnBvMXzpU4PMc2dA5Bx0j+nsUm079bXIDWuwRvdlAUebXRXjk4JEvl1eq4Z//a2l8MtfHsN1lol7KMBSssa2+kD3zQwvhzByYrg7/OznR8IXPjEUQ/BvtZZ4QPqUFEgKJAWSAkmBpEBS4CFVIAGsh3Rj07KSAkmBpEBS4MFVQJeS5WTvnCtHkGV+VA81Y/vGcmHfBG4cSvoEG11drZgt5UrXyXEyw6m1laGU62boIqg5TRdAYc8aji5dXZbl8SOwyZB08qRwFZldle1p51cJtbboOriyWgc6UbJHuWG12gNQ6Qk1/oIQP23hGtIxtmdPPgxtZuN8ltYIkYLCDACmunj+ySO9MWy8RtK6AEooVgXaxOx3vs9aRgh8G8JdZU7VAqWZgiHLEPtwR1V48elLtVhWWOC1Zlpdna+GyzNdMSTe8kfLEXVrHSN36cuAJx1f/lwC6A2TwyQs22l4riIwaj+69gG+RkqVsG/fII6nHFlbdGYEPm0AEQVUvrZE6V8nv0r45IdDaKUza4g158j1qvF6/VnIGb/CtkIdmnX6Yi28fmojHCP/yw9VFNJl2N/tEMtzqn/hWnmgUMsy0DSSAkmBpEBSICmQFEgKPKoKJID1qO58WndSICmQFEgK7FoFypuN6JSqU8ZnqZmZSuZVDQBURoezYYRcphNHi9HFsxeY1XFZVcmDEiZZzibw2D6EVU0olWBHB0DB4IQAAEAASURBVE+VyCohksHrvt6xWWl/U8gS8s63PYCsHuBOnR9WYjkhTi+cQct0KtSdpZtKQFTko0753Ui/nQM5ENDlV0FVl+4uKM7+iUw4tr8vnLxYDu+cqQDE6NBHWZ5d6o7sL8TyQYPLhURgsxjIvrjSPscieVBOModza4DXO8sWYO0sLrJNAuFPX95g3ZqdtsigqocGGvyJL4zQ3bANm27UIi722ieBlHNvESTfauCKqgK8CL4fxxHmcUsAN0sWdaZFmLXZHSboyNjRvHMufzb9SWj34one8C/+cCnM4zyLA3jlsd2soUHG1bnLFWAX0I59tFzQNU9N5sI59nl7rtZ+3GudsPrOddLXpEBSICmQFEgKJAWSAo+qAglgPao7n9adFEgKJAWSArtWgTXdVMAOXVEzdAa0s51dA9c3KcsDFs3iTspfDuFAsxhdOTqM4uAYjzMTqucGamNVnzlO6+RZWaKHdyq6pQQrvtSvlv0ZuA4TAqq0P8yCypHVVAA4WRJn7lWFUkGv0V80KF0HFfCLOVc5SIeVOVBwK8rgAo4qHVSCNRxfOJkO4HQqkx2lq6wfwLR/IhudV56vU0UnwtKJZT6VJXl5XFeex1I/IVeFLKwW81hHl1n0MftKXscpOW8rXJmpAIMq4SgOpzyTM/vrVg4s3VSWMJo5lQfcZUgEO7zXDK72dSzjXMDdFXPB+D7D9WuNHK8pxPnqbOOlbSho+SXzmiJ4/S/8wkT4h/98DvBHF0XmR0RXnP/4iCWgln+yr4DK5TU7JWai4ysPWJxfvNY9EoA2yuO7YaiBGnqP6Ba7lZa7Ya5pDkmBpEBSICmQFEgKPLwK7I6/jB5efdPKkgJJgaRAUiAp8EMrkKWczM585jsZLB5hEo/FznecDdYTYZZEQdfUcH/7EjncUTqBLGW7cQh+RnAWXZmrRSDkazA0xXMLjixxE145/CqsINcd8NQCXpiDhXuIboNlHEo+KdTK5ywf7A5N8qh6gEyW+lkG182EiYSPsKsF/BgDxqwBreaATdbUFekk2MVrLInrBYIJ0CwjdH3rQC6h0CLd/ArAsULeTKzuCKbsGKjPSrikMymPO6zWxLUU8VX7Mz/E861u1Dkmz4ew7GY94kKvfWpDrB7cZZlQLRtmL+Azs8vSwEY4dZHSyyoCMTpgq49re6GOY0rdBwBydV1tHPdFyhifPlYK/8fvzIc3z6zH1xXJEqsA5TYpVbRM1H31a2cIIq/DyM6D9/mrjj418J5zVKkOdX8EjGkkBZICSYGkQFIgKZAUuJcKJIB1L9VO10oKJAWSAkmBpMBtKDC1Jxdef7+HErsMQeh1nEhdlLHhfgEaCXsMWO8tdcdyuizgxGHQuu4swYKwQaeUrh+hkBBH15MeJUvjpild209mVXMaZw3uKKEKFYHRwdR9DWp5Vo/XcaPbyS6IPdiq6vUqgeu5MDlO3hVzWsEFlQEoCaMMVpedFYl2arWAU0CeLsDbFnNYwsU0AxFbWKkGw9hFXEK6g/tqYRSwdhBHUwU6Yic+hyBLd1e7Sx9h6cxhEaCnW2pshBwuOvMZdD49X2NOpEtxDYFc2yFEnhZrFDb5Idi7nWFg/grlkUW6M3qc57Pk8eIM1AYyJljUNSbQuThbx02mm6o9dF6VgYnqr7vMksMNoNc+dLq6kA11fnZfBGGLayHsxYlm7pVuuN08LD113rrILKUUNOrGKlJ6eWMZ5W5eR5rbnVFAeOk9IeR1//3/nJ2A+Z25WjpLUiApkBRICiQFPqzALv+z6cOTTT8lBZICSYGkQFLgUVBAF87nnu8Pf/TaWhgbJHS9bP5TC5iDI4o3kPlChhK2fDi8rwAgsZufsKWd92Re1mWAi7lVvfmecHwKUAKEEkAU+Hl8JJA51YzHHN6bDW+f2cSVVQ+bQAnq5+IQqhjuHtOooBfCmejKYg6FLssQ62F6LhOef7wUlglcF8pY9rhe4bzAIgGOHQSFHV53k8eXeN3F2UqYATiVcXLpVMqyIDOr7FT4Jt0RR4d6wt6RHHCupw2uwFyri43w219fiWWBTeYyDjz70kuD4SdeHAiHgHBl5v31V1fJrmpEDfrRTjh0CCBWoiTPOfQCW37Q2GD+sWSxbbSK0O8K5XxLK8Ao1icErHSRI0ZW1shgjvB45j2WiaCrc+4abjmSw+IbfDUzi8vw/WEC6suUf5o51sUaCoBIOzT2s04D5HfrcA2CxIszhu8bQW9nyZ5wYDIPPHXW3hu40gCqwtU0Hm4FhFdCamH2//eNhfDN729EkPXHPjMQ/tIv7Xm4F59WlxRICiQFkgK7QoEEsHbFNqRJJAWSAkmBpEBS4MMK2KFOSHVxthbO0D3w7TNrwJQWbocQThwuhscP9Yb9uHuGyIXqjMuUB75/vhyhkY9txpK/ZnjsYBG4A1rhWMvlJkbyuGjaYe4DOJXeOrWJS6hKOaDuCkPLcVDxUafGrZt6uCzOqyJh44IwXVnma/km9vx0uxxxBsikK2OJN7dFQMYoLq+BPqGXZYbt/KpWi2uwlk3LDAE5sp56qxEuzGDlgIyskptlttJluv8d1J3keXBo/d53Vwk9L0eIJkCZWaqFr35rJUwBrz79VG/4mc8MhxeO91GmV47rNWD9c88OxrkK0zpdAjsa7fRVh9Es510n22uNsHqhmk6yaQCWbrduhNMZJjxs4DgrFoBlIRffzBsq33Gg2PXQ4evgXW2Qh17jQLlVc8N8kEyxsdFMzALTTWaI/G4c5ozpSHv3QiWcwYXmmnTUDdd72D+dZl0439gP7puV9VYoobWNBVxTGg+nApvcE+bY/ff/99Xw3TfXuce5nxn/4P+thLcA4f/NXz/ycC48rSopkBRICiQFdo0C6a+MXbMVaSJJgaRAUiApkBT4sALmPO0D6FgleGQfeU68YRRA6XjJQpK2wytL1t4/XwEmUOrHAXYAtAyujPtpgTypScCOgMjXDfZ1hz2juTC7WItg4thBws4BEOvrtXB2tl0+53vTYtbwc0rihDE8YM5VLtcFhALqAHoWlzkeiLGCO0dnRgUAZtZWGUg1tUVnwQFKC3EaeYxOKV1depQw9rQzlfjBN8Rzy/UYVG9J0irn3eRjL+V5Cyvd5GbRVZADREOux5B2S/jsZPj55/pZC+WTXHM/XfwsH7S0zbGOo+rCdBVnGd0OeWzfWDYCulWuZ06W4EkHlENQI7RzWCanq8z1bgCreuio6LV1Gqmd2vYW6FKInr5G11bnPJ1rC7/asNCOhj0RXhnIristzxwn2dMnDhVi+VW86C0+ma+1jharG0K1anRz7WPfhn+EcHedcB3QdovLXX/YroynLlbCIqWU33t7HddcHccc1BEldAcewrnnvASLJ+mouOreo81QKRc++0Ix/JmfHqekNP2JeV3Qh+QbAezV+Wp45Z0P4JVL4zYN3313I5y+UA3HDhJwl0ZSICmQFEgKJAXukgLpr4u7JGw6bVIgKZAUSAo8eAoIMRbIZ9oPiMkCRXbL2Knr2zWzT5yigGcJB1MVQhRhEw4ZQUwvrikdRLqvfL0AY7A/G0v6dDDVoDJtOBXCQXK3rs43AT4ekyHbqhWI1YL10I0PPQRPOoz8ua/g18DrKT00lJzrVoiJKgjNeNzHFlfrYRIIVa8bdE4+FyBMsOOBzsWPeG1OvLomQMKxxUeZl1yaq8SSxqP7i/F8XRzooQIjs7OauKFauLd84yzQy7rAbUModuoS5wDeOVbXQ7C0cnSwJ77exyyNGx/OUQ5IEDxB9YK3Tcv8WGvBVcY5tsPidRzNAuvMChujhNHzDAJydCk5B8Gb8ErHl8Pv7RypE63WyPK6rTAPpMtHV1kmPH20N4wN/+A/wYREOs7OXK7iiHMHdDzVwvGDpXAQYHc7Q7DkWp2n+1cis6gIyPyocQFn3exilTmb9VWhfLMaAZ+lqjr0DOTfT1bYGTTWrSYk9R5ZYK5XOO7NU9Xw678yEU4cKn3UZdJzD5gCuhnNnGvy+3Dj2OL+fPmtZQDW5I1PpZ+TAkmBpEBSIClwxxT4wX893bFLpRMlBZICSYGkQFJg9yrwrdfXwivvbYRFXEnfOTkTXnyiN3z6mT5Ahujk/o0s/6ZuO4/abxp1QVjOZXi7mTQCHF1AHTilO8khHNJJU8xnADVZnErtbCbzqfqKgCWOPX/FDnsb0dEkXGoAcjI4jgb7cX1l6sAcAQ2ACAcVvCi6bAxyzxV1YdEhEeAkFKkSXu5r1wBFWXKyekiCd449PZSbHRCktcIBANnQuQzdBQFqrKkBPIv5SVzYAHlVproulqc1WJPQxWD6sZEspYnCKmFUG5zpCnv8cO91GOV6t485oIqwqTOEN/PsK3HuYZLyPUcDIXRoVWucHwdZJ8Op5WQY6nmIEs6vf38tvH9hsw2rAFutZj28dnKTbLFidF6p/xAwa/vogCzBVmFfMewl4yuLFpkeOkYSWH8Db9t+6PXvdV+VAWRXySfrwCufFLZN44IZJEze7ozuveJ5LZ1mzkdoVQcoCQUFS7rBHAJAdXVtArudhgHzr59bC5fR5uzVWjjJ2nW9OWeD+c0D84KCsDLfey3zw9RYxtZs1bmnWuF/+Ccz4dd/aSI8fbx3p8ukxx5ABSwhfYzfO7tnXuOp3nrx/vN3/RhgNY2kQFIgKZAUSArcTQU+/BfX3bxSOndSICmQFEgKJAV2qQLvkRv1nbdXecPfhh51rFjffXsDp00Gt0vxvs5agDZAZ7tVHD3CC7OULEWzXG4Z15XQImYrMUtdOWuUAa6SSQRSAFRkwhGytIbJyfKNpmWCDrOLXjtZDr//3SUcPhWcTziceKqXsi+zjMhSj0Ck0QRYcM0Mb04HBvE+CTIoXczj4ClrucIPpQtJEGU+TszOQsLubpxdzoHvi4WtMIHTSeD2BYLXX3tvlZI9868oTaQrYXeX2VJAMI4Rojk6MMbw9Rce6yX7q0FZXxte9QPiPvdsf3iBAPlbDcsa8XNdf1oHlXroUOPsEd6Z2aXDaG2j3gaCaNpftNwQlxrXMPzdcYlA/ArHWTrovLq6W+HidCVcJs/rOKWXutx2GkKsDsja6fkf/JgZU2jDx41Dh9sluiCaQyaQ8h7ZQ6B81BmXl8c5NnBGVXF/9dGxsgPofFzolDPMbNtQj0tAq2+/XQ2X5hsAv0aYJVNNqOmvhVvTAm4KNM1Aw6fH43ao5HmkNjNNxb0XNgCaZ8nN+p9+cy589tmN8JOfGeJ3SXB3bYO3XTd9++AowLbjqiuEZx7vC6+/C6128Ji7aonzJ5+Iyf7x4fQpKZAUSAokBZICd0OBBLDuhqrpnEmBpEBS4CFQ4G//z5fD++fKvClt5yX9+p8bD4/tKTwEK7t5CSfJjrpxtHD8vE+A9f0EWItkMf3bV1dwCtUAViEcBUbtJT+pG5Cgw0rEYh5VTjAFOdBV8yzAxxwsIc7+SeAVMMaSn8tz7ZB2gcf8cjP8099bAIJUgTmcBPuMAKvZ4KAS5XAAquZWBnjFz7w93YJEbeHk6TUwHkDRRkl8K7zgFRiurg/f5OrQsOawUgeKAFp0NeWBF4PkQR3ZX4pQrAm02qK80XNNU4q4KtRiLlsszFK1AUodwWB0JSyEX/6pEWBdK7p/xgmgf+4xMrsANbcavcx/YduT8DdgHqVz1zr+CYDKABvh2whzKjI3Adw6pXEWK+r8EmCtCreYlBlbnTB4XU+imlVcRzqv7haUUUehk46quEfX1uPjNQCU5XorQM1Yxuh8cJHphBJi9eKOag/VbXdX3A6wrj15/Ysg1LyruSXy0ih1XF51ha0A/4xQ0T2WiVEEGjCRBbgVwAyoWqbUlFLReFN0zsaldWw1mhlAZS28e55zMemvvDQURiipvFt6dS6fvt5dBdy/v/M3job/+n+/FF5+A/iNNfPpY73hP/v3DqS9vbvSp7MnBZICSYGkAAokgJVug6RAUiApkBS4SYG//fcu0vVu49rjTdwm9fDf/v16+B//06M3vfZheOBWZYK3evxerHmZYOzf+Nfz4SxB2TUsLuvlevgGZY6ffKo/PH+sGHoJIDeYPCvkANA0eWNpSV+GerHRwVx0//Tj3HJYKqfrxnynK5SffZdSyTNXNgAhlJnprgE6CCZ00sBvgEcZMrGyuLPIQaITX40XFQApNWCZ+VM9DaAYFEOWU6Kc0OwrgVq53KBbYXfsMtgEYsSyQ2DRHOV7TSBZYSnDXHE54VpqZ4yRp4Xr65njuXD60mbMV3ICwwM5QB2h84C5GuHhe8byYYpoHWFNlTnPMifD4fuuhbDHRW77NMmxi6sZgFS7nFJ4dQB3Wgkw5jDQfBMDWZ7SSN1CGaxf5owVc4AXJDuyrxBmuIYlUyOU/Fl614rioBM6q+skNrW7DWPM0Nozmuce2AQomctF6R/70MNH1BY9Ok4zAZNh/YK93mI7H8vXVqrt9W6T57oTz8fU1HJSXVtuv/eDILQBYHRfe9ClARhjq0M3OgjyCtQTOrcXe/viPr5NB8hYmsj8uoCr3hfZrBAMrgrfMrfrLGHvhXyR8Pyb//Q0HN9SSPdFcCiAu9vaMrW7NnT8dYDnXbvIfTyxEPVv/tqBOANLXe/n/0/eRxnSpZMCSYGkQFLgPihw818R92ES6ZJJgaRAUiApsLsUOHnB8rAPjxVK077x6mr4McrAHrZx9ACQ4AoJ3ttGN7aiJ47cv/JByxovTG9GeLWGo2q1LGSoh++8sQpMaIYp3HCWDh4Ctvhm3zJBoZfOIjOOHLy3BE60IrzyZ8sPT+IqO0uHuQijgBWW+UXM5SF85LPkJulOAoQIMCzjy8VSMf+rF+VqfF+Hcui6KhMy3qL8sIWTqq9At0H/quAcdkmsEt4uGPPNbg1w5nN1IJCh8rp9dJE5bxK2wnBvLnzpk0Ox8113j10Ss9GFpcMJXkZZJCWErM2A/RzzE9LEQHLmNrQNiAhffEMtxHv6WCnM4lKy0tE1CfUa6LWOa0mfla6mMsa76CLDYlRrdFM+yTyZkw6yNkQJ4SVy0H7/u6toLwyzu2N32D9eIIPr1iWMvPCODPfx+FQBoEg3RtZugH0/5YA6nE5drF6HV15Mnbt5hCVeH7q3+oCFQj+HrykB5XSUdQa3RxyW/um0KwL51qscB6RrccAqrjQBXxcbYbfFMcDVY4fy6J6NDr8//eXRCB9/8/cXw1lC3XVk2a3SUswiH7rpHJYo6nAbuKHKzBLJNe8jJt6+V3RvkRXWj9/r2n0cT/AAfJpZaIT/8h9eDldoQiDA+vHnB8Jf+zN7HoCZ/+hTTPDqR9cuHZkUSAokBZICP7wCCWD98JqlI5ICSYGkwEOvQCdT6caF6lZ5GMeTR0oEeTfD6++tBZNdcqSMf/LJUsx1uV/rXccApzNGd4rOKUPJ13A4zS3xQUc8S8U+/fRAhFSFfCmQuU4XPeEVIIMPHVc6WgQWIgThzdIK5YMcL+gwR6uTryT48bgMB/b1CnB07tDVkHMWCUwnL5yCMp06PWGFcrPNCjlcPGb+Vh5AYbminQJrdSARQKJFkLsOMWGPjhpzrnq4QBdwyEwtS+CEW0UcVovMaXI0hCFK+Z57ohTLHXXj6B7KckwWh5dAzW53gg1hTI48LjOoXMMnCdv32sI59XIIQsymmhih3PIDVkOEO+HjOIoEZzqGovMHd5FrMCy/RYr8cF/7AN1fPv/00RJ69YR3zqyHKuc/NJkNP/7CYCyzjBe7y5/cyz2sw4/OmFuqc/0Gc/5gzbrwSkDEMXLbtg+dZePDPdE5JhBTm+3DfYxZWaxV4DlGh8UK5X+6uQo41PZRsup+We45AiTdM5qNUMwcNOGUoOsEvz9/bTQX3jq9Ef7Vt9coZ6xxD23RaZFmAOyhQDHLfVSjpBRv1vbLR6i1TPmj92udvXUvDLqv1u2WeMNkP3Tk7vrBe/Zv/HfnwuLyB+XIv/31+QhK/6NfvbcQSyj7WwDFi5QeDwOBf+6z/WFi4v7B+N21U2k2SYGkQFIgKfAgK/Dhv3Ie5JWkuScFkgJJgaTAHVNgL/k2l2bb5Vedkxbz2fD5Zx/eN0Gff64/fIJg8LdPVcOzJ8YogYq+pM7y7/nXyVGgA8BBR5Ed/VYFLpSDOXyzfxmXx8tv8GYfdw44KAxBmaLjie8tjRPgtIHOVsx+WgRICI98zLLDPjsL6jySKQCwCoAuc40mh4VJeK14E1yzi2Ef4eY8p5NrhTmskI/k9QUfurfMSCsBJ/p6M+HoVB6I1gZuQtAGxywDlrq6eqIDq1rrCmtYfQQllqllmIcuL8PlBSEHgCV/8MoawenXIADQZGoP5ZKUlAlTKhyoq8hMK6GKpYlLa3T4YzIdeLUBSKhSFrcMkNTBZZ7XB5lQUb6YFWUm2IXLNbKdWvFN/lM4trzv6xXDyYFkzEvH0Trw7KljeVxHZEuhg2WLPnc/h80FpuiQaCdI98hQ9X6caJaUTgCS3BXLQ4VVgriPKmdTS6o0Y1fBNaCpwOxZtLCbpa6sCnByg/JVwaX8SX1zlFrm0dx9d3iNIaDTJ57sj66sP3xlFbcc1jcmICwdxDVm98K1DTthAjevZ3TRHRIYdxXn0pun18MS2Wy6wx6jccJPfLI/7nGV9RX4XRwgy203j9/51sqH4FVnrt94bTXcS4Dl79Z/8fcuU2b7AUh7HbD4139tf9jP71caSYGkQFIgKZAUeJAVSADrQd69NPekQFIgKXCXFPgPf/Vg+K/+l0u8IWuX1QmvfvVnR0Jf38P9BsjcnhEcOPcbXrmtTx8thieP9YVvv74CYGheBzQCB7gOeUSUBG7Ww0nKASs4tMYp7Tq8vwBcETDRgRCXkWCBV+NwyrGXjeiyKuSBFXWynUayIV+mvAywUCSHagpn0QBwJtNdCb2AiVXKugRoDc49CEyzgHB5bROIVccp0wqbQJ4stKOXSrp+INqRvYVwYDwLAKH0iwkKftZxjE1W6YxoNpbnBF7E8kNI2xDHCIIsN7NMkVil8O231sleW4+ATDhmmaLusBN0+yvmgWotXFjAFtdnKSJsI8wt1MMwQMf16piqsB6Hbi0BWpl39J6704FxbaMVvvnGesyV0iWkQrEj4WIzDNIxcYYOfDOra9E5dHR/npK5bARuNzqX4kXu0ycdcnac1G01a74YOuiGE9i1ywMtdby9ydk50i6XZcoG7Uw4RrbXfjrK1Wo5wCcOOQDk26fQmUww87JIvGf/m/G6Qr/OEDAWqBO1lPX4VDG8e7Yczl8BoiBcuwui0I8MNjob7p/QmYXDkHurTKfEb72xTJA/wIt72pI03XbzK7XwOB3vLCEtcE0daIeAdrt1zAFEdxpC43s5fuePVj4Er7y2jsnf+upC+A/+nb33cirpWkmBpEBSICmQFLjjCiSAdcclTSdMCiQFkgIPvgIHJ3vC3/mbh8I3X1sJNBILX3i+iOvi4YZXu23XfCP/az8/BhTKhf/nDxcjBDIcWpoCO4plc1qn7DhYXKJ8T2gFDDIjyZB0j7dET3ijk2YcAMBDMZD8FFlFZkgV88CrXB7XDDAEADazUA3rZE1dwAHFJXjekjOCzgEXzWY33QuBJYR7x5JBzlXjVbHzHV0G7TAYQ+P53qyrXkLWBRWCqgb5UlWgxxVgk0HqghC7/gmjhGQruKXeOV2JMM5rWQqps8iELMfsUjaWtm1strvq+biuqyFcXw30EILk6By4hkNMcKVLKw4XzDCDK8/tq37nCLRfAo4YNi4kEy90Q3FOX9rg2Gx47X00KM9H8DJJgPpXPj0YXjjRG8/j8brcnPNuGEKrw2Sg6bq5ttQfelo6rjzWjLJeMqvqgCzXF0tDuaeaaDfSjyuN6+iuEzqOAz/tfLlM3pXHDt7gjvKeMz9O3d2bMnuqk5ArRagZr8d53K9FYJXloJaeundee5OstbfObFECynV43Sbz8Fy6sCyJ243jM58aCf/iG0vcU22A2pnjOAD0Xo6LVzvNNz581fnVewvSPnz19FNSICmQFEgKJAXujAK786+AO7O2dJakQFIgKZAU+JgKfO75wY95hnT4x1Xgyy8NxFyqv//PWmF6sRq7xQl3BA89lOZZYrW02oxd9Szt24cL6iBOFcvBDgCQdOvUcCXpzBIwHdxXBBgBE3g/292TDZNAArONNjh2Esh1da4rlJergC4BF9lFOGtmcSVVsfnoJonldTxmcRiX5xxmbAE1AGC6lXQ6Na9BJDOoPIfDEr/jU81wGFh2+lI1vHd+I5QhcdGJw+tnCFwv0zWwDCQxiF5Ip0NqjdI1Th9L1Cx39DnzvMzrqkLmhFq6qoYGKLNcE+gRIs4lhSzR0uNM0cChPrUYxo6rBzAi2HM9ZbLdNnCJnaHj4wLz6KMcU33PXG6G3/t2iwD4bCyxE7AIYnST6Vr6UaFRnMwd/PRx5qHOwkSzxYREDvdzkPUtGciOmMJO75F6qe1m023lvlXRwxy1KtDLzKzzM4Ao9lloeexAMZYuup/+s4lBaXbRclLzyQCPhP8bqH+BnCZpqQC2jraWnjbZ8ir7JOD0vvUFG8CuNeDqbgVYLx7Jhc9QhvytN9bClrY2RrGQDX/1V/bF7+/Vp3Ggazh5M8QyFD+NpEBSICmQFEgKPOgKJID1oO9gmn9SICmQFEgKPPQKHAY6/cyPD4V/883FmEW0VgYgQI8GKXccxOmU5z2rQendwIYV3uRLCcxC0mHk0NVieV3MtAIw2UXuACWDPp8B5MwvNXHT1MLscp1OcpTpARI8plTCiYObiTx3nFw6rwBYfPicPw/g9tpLR77HDhVjV7om8KLZA+ThvJa1+foOwNIVlSUc/4kjPUCgLrKrGhFW6IzqwtZk2ZhrqHNMG6Mwcb7J6CIT1vGh00dAYra3kEQsZQC5pWjTuMeuztcBH3SwA7Z0Q6ByuUYYJZup03VP8FUwWJ41G1DvnCrAL8HIRrUNwCp8rUNQDD/3+Uu4g15+azW8eKIv5oCppwHwoJWbnEc+9yANg8ctn7TUdItN83YR6HVK/nIIf2CCTpDswypdSNfphBn3hk/uK6Y3oBSdMd8uh/fPleP+DZDFlUPjc1dqEfQJGdVR+FclRCumYnHCJt0so2OOc9FcM95zntx9blIe6jG6Cy1J1QVoyP6W9GsXj7/1lw+EPyLz6uuvbfC70RV+6Yvj/H7cW3D0J78wEr5HKe4SQfqdkcde9ye+lP5jREeP9DUpkBRICiQFHlwFEsB6cPcuzTwpkBRICiQFHhEFzJP/sWf76YCXiaV2K0CXFsBB2NKDS0kw0wUU0mhkPpbwyJyiddxJsVOgZVi4bBxmR+ku4giOA2DgsvneOxvhFK4ouxvWYpSPnQrJRgKQjeA+MkTe0kTxgWBIeKUDZ6A/R8h6PuYtXSEDSJBU4nXDQIxeHDyWF/bh2tFJYyaWcMTz5iBQgzhCOE2ESb5O59Yg5wMf0WnRjoM6WAx2zwNW2vOvst41XDtrOLUajRxzwzkGCMtxommCwHV+bVHCtbzMfAkKL/Ghw8wQc91Xunh0Y9lZ7+wVyxR1+lheCAykBHLNfC/ZlPlanMvyNSHflZk6gG6TMjs641G251zncGoN9j24TQ10klne55ALeZ8YAj/dQjd09x4ZtvyN1x3G0XfmEjvDRqqxgfFDfJTZD0HVuavlCKd02QmdBvqyuKvauWdF9mCT12XQ3fuyD7Azw/5euFoL753bCO+cq7AvgiscgFxfKGp54uQIe2s+G8H/7pnQc/iGUsU4+V326fPPDwQ/7tfwXv9P/uJU+M2vzcdwfKHvz31xKBwa3735YfdLq3TdpEBSICmQFHjwFEgA6y7tmX8Ytjr/6fsuXeNun7Z5zQJ/t6+zW8+f1t9+Y7Nb9+duz+tR3f92Tg0A5BH9/XfdarDb1m8gtw4qwYsfjiUybd4mKHsB15Sh6bChUAFIzPD4pW8uEbAeAAKZ8Kmn+mIpYalgmDMv8n+QhBZrtbzrHODqykyVMkE68m37tRd3ra4LcZqxXK4dhI57in903owO5jgvpXUEn3fhnLKM0Tka9m4Hw60uHEq9XeH9841waZpUI67LgzGwXchVyG4RCo+bK3a54zh0t0ytRA7TUG8WNxgQjfNeJI/r7NVqOANIW99gjkASHVnmZAmxNvlZV9U6YW053rzr4LE+cgtBhFmmZtUpR1uO8wNaschhYFUhWwjrwLo9gBLdPTqCphcom1syJ8xztF1nrrvOMRfRSNeR5XOCs61WhtK5xkd2+GufZXd+FibZHMCB9NEtt0nLQfXyXhFSCvFW13XcbRGgngVi4sJDb8sLdWmdvFjlWLoTUsYau1YCA3XdCVF6EC7DfucQM8/+CMhq5KfZlfLK3GZ4+U1dW5GWtgUColWnG+Gzz/eG4/twD2414z4L0rwnpiYL8ffybv5fk3+37cbf/7ZAt/+5CFf98z8/+qEDbuf/01y/H7fz2g+d/CH5Ia0/7b+38qN8/+t2fVTX33nv/iCvv8f/svgIjK7V1dVrf6Y9Aqu9h0us44e/fPnyPbzinb1UpVKhpTp/wT6iI60/7f+jev/7L/BqtfrI/v67/lqtRkne7nMrCBMqxgXxb22hSgaocGF2K1yeE+TY9a9FBpF5RjiSAAiWyekgGhvoCs8fzwGhzHVq/yt/dWMrXOS4heVWOHm5RUe+Bu6rnf8PX5PR1DjnM/eKi/cBpfqAEXvpTGiuv66vEnLRoBB4QXg7/+oQbhRwWa2StXV2GigALelmTrpwRoaYF19fP4XTatUyvhD2j3eHicHuMD5MMDhQ68p8i7kR8I3TTHeOrrKtuOZ2iRmMhZB4Ar37QphhDZs0uxOyeJ6RAdbJa/dwrlFA05OH2m6iqwvt/C5XKaDS0RNDy5nzpTnKGXlsfqUVXj9ZJ0+M/K8MoeH9Wc7JuuA8A0U1bWvk+ibo1nf8AC6yCMx21m63P7qE1oIsSwN1x713AahJWeFzR7O4rtpOKO8391V93KuYqQZ09HflNBlh63SrvDDTxHmFI1CAhavN+2MQ2DUxIrQib4z7w4YA7tMK13r7dD28drp97Rs1OrG/O3zySUpF0dx7bQhY5nkG2esMYg/hCrxbwzcufjyqDStcuxAn27nR75bQu/S8j/r6Gzgs/b1+VPff9TsylLo/isP3rgKsR3n9OswfVAjk36179+59JG7dR/M39B5srf/nf/jw4XtwpbtzidOnTz/Q8/+4qqT1p/1/kH9/P8797x+vZ8+efWR//30Dc/Hixdtev+DIoePkfozDh7co/duM7qdFQrQrrfXYKS5LCdZ1sIJdqdQ/EKb2F/jDzFIsYBeldD0Fwsr762FxcwOAs87023+8b1+HcMrQ7PHRYnTRVFjvAFDBjoYnzL2iC6DOKoPVB8yigoYYxo0hJ1yYa3eXi6VkdEXMbXXHcPT5lRBeO1Vhnrp5LIXkyq2e8MTR4XB0qkCnQ4Lcm5thFdpR5A2FXd02cI5VACs60TKsRxBm+eRrOHY2AGfR3cXEpxdC+PSzxfDM8WIMpD9E6duJo70x56mV3eb2ubbIAmHslsIN05lwhoD80ZEG2VqroSc7wJpzgDlKFgEzwkGhYQvtGkAa87FGxgphfLIY+libLqEHcezBMWew+vzZzfC//c400JC+kmzId94N4ed/Yiz88heHKQe89Z+K+b4KHQSrobefAPwrVeBjPTqwRgZy4Sj3m90t7Vxo2L8dMX1zNLdYC2+eXQKY3Rw0robLdZoBjI/idAt0nswCI7kJ+d84DiydX0N3sQthuVwOKysrj8ybgBvv2fX19bCxsREmJydvfOqR+Jn/qB//A874ONT6ERze+0Kc0dEPu/ceFSmWsN/678eRkZFHZckfWufCwkKEN0NDQx96/FH5YW5uLv7Hy4GB+1cC/qho/XHXeeu/Sj7umdPxSYGkQFIgKZAUeEgV0A11nu5p3yXcW5Dy1JHe8OSx4vVMJ8uv7uTQHWMw+o2gTJfVk0dKMd/pIllEZlgtdNWjm6ZzfbwzMZw7jwOmj2yjCiVehmzrQNrEWjSMm8h8qxwumXb+VftIAVgvLppnjvfhROI4ANUQ4KYfYGNI+grZU1MEwXuulTVzigKakE2FNouUNdqVbpUcJN8QrNHhr5in7I5/zlCyuLSCy01HF8d0s7YlXvv++c0wSQc1j28Atrp1OyGjiLDJ4ps8VgWCOTK5LdbaCqu4s7YP1/ZtArSfBFrtB3g8frAQ853gJtdL4ra/3tI2nWmH917L0ypXw56hzXDs8DBzboVF5pnpRiNAyibrF8bZoc9sLUHP+asVNM2EQ5S8dTodbj//bv/e8HZvrH/8u7N0X7TzpM4qgvFZ++9+Yyk8ebgQXnqqP2q401qmJvPc/yaUdYXHuednlzPcoyGWuZo9VgFmCpx60WuQr8vcM7OLNAZoV8HudMpQYk57Rtl89myT47vJxnLMLxGszz3j/NzPNJICSYGkQFIgKZAUePQUSADr0dvztOKkQFIgKZAU+JgKfO/tjfBPfm8hlMldkrB88/WV8Pzj/eGXvzwSHULm/fhxJ8bymuVZQClKqgRYhjLrpOrmXbxOIF1Sgog949nQi6NopbsRs6g6sGuAPKk+OhUKnjyuTysRwzI4wdiGDiMgXDdA6xwuGnKzg6HxA6Vs+MInB8JnnimFN05XQiODw0p3FfBIfpDFCVXB1GRuFYiJwGi6zjEP4VMsceRaeeZT5YRCrAqwjGQkygqBWhxBpVIMdHcBBtIvAIsaOLjsPicArOG4MmBdlxt96ChBtARSrkFJIo+s8/xOo8KJX3m3HD77TF/bvcOLPN8AbjE77pmhpWZ5YRxliIInXUYDlKhRPRpmZ9uwZbCfDDFAitlXa4CrHuoSDTAvUapYxj0moOuKKQzVCOqeOlaKc99pTrv1MWHQ2UsVcq7apSOWiPpYk09lAu3fo7Pgiyd6417vtAb36sThEse3A/otExRomq22yb73QEHNxPJ1DjPEvEcPTOTD26VK7EK4/bxC00N7MxGK9XJ/dgGs4n3CHpm/5T1vM4I79bu1/drp+6RAUiApkBRICiQFdr8CCWDt/j1KM0wKJAWSAkmBXaSA0Oef/f5CmF0g0Ic3+77pFgy9dnI9vPR0H2/A8zFHyPypH8WJFR1IgJIe3rTr7pql5MrSNccysGlmvh6muIbXNa9oAAizSOnWu+c2o6MJXhABUZOKwNGhHPlXRZxIvJ45Ojp5WXaLE+Qc3JcPE3R/e+Hx3jAPhDp9eSkMQXOO7i/hmummRKwec3HMN4qle5SdCSj6KedyXn61K9+hveRV8bjliTpnDEVvtoA8EAiRlzlKvq4PWLQOHHG4RoeaFqBm5maBqcJBOhvO0uVvhY6DUUPARQ+ZVAKMkiCOw/IZ5iHVu2F4xlXS2b/+ykoMW983novXzQve2peLRwiuioSDf9QQvBygDG5+iU56aC2ccw+q19LuXY/nXNmoR4B3EEfSgzS81wo4yNT1w8P7TQCZYY+40T5ieB8air992GHSj86wA6Qgqpe96zrURUlmPcLEb7y2FsszdRf6+/LsY73h8L5CBJotykotI+zi+kKx3mImwiszt3R0eS6/ppEUSAokBZICSYGkwKOjQAJYj85ep5UmBZICSYGkwB1Q4J0z5XDu8mZ0lng6GADuJtxQfH95rhoBlo6jOsBmjiBy3S053EOTo3bY++BN/U5T0cmyFrvr2cl2C+dVM0IdXytsaHf5A9DgCDKbSjDwztlKeO/8BjDLrm2Etg9RxtWdi2DpOJlSe4BTHZDmOe0CZ16W3QuFUnYjLBaysdzusYO58Nj+Stg/NRFWVxvhNPlarkW8sUJ3Q8PRh3CA2YmuKoNi7TpizIXyVfCOkMeRJfCpN7uZr137KL8DxBVx0+wZ4/FagQB3OtdxXaGarq1+XGKfea6PYPgsZWbNcOhAnufpdpgpcx1KHnlNlrVtAuV0dwnnmszh/Uu6u1TngyFQ8cFpANhbZypogdOLT33AjjztEwV4Mhlda7dT9ufeCcEWVgR5ZDpRZ1kmk8tw8i3Ii24gnUYCLgPR4/U/mM6u/k4dpvbQ2RLwdvYi7jT3ms1WlyFyrF56Eheben7MEUsVOYegUlD13AndcbgU2ZNF7rMtuhMe2pcFetnpECgIxJzn/rQcscieHZjIxDLUZX4f6DET3j6zHq7ON8MEjQT+zE8O4z58dJvOfMytSYcnBZICSYGkQFLggVIgAawHarvSZJMCSYGkQFLgfirgG/D3cDrpJjK4XOeKIdNrZUqncIsMxnK69gxPXtyMbqQ6AMZyvvllgsrJq/IN+o1DsLSw0gxzQBfhVCdnSdiU4dxmXS3xRt+8Icuw8rkMcKhGnlSFErBqWIHs9AAfcsAUQ83Nm+ol20lItLDcBbQhBHuEYzjekPQ3T22Et06XCT2vggqAXsPZ8DOfHw5PHc6FrapttC0lM7uqGV6lHO/STCUsAZaoMsRt0x2O7esNj5OPpAHqbc6jC0ugtARgyLPWob4swew60MzaandGLDOvC2Rklas1QoIz5F4Jz3oAbtnwqz83TpZXIYITpOAYOic+3hfGh7op38PNxWOYn8LcQoNyNJxcaNCHe6rGyc9eou5v2xCK6Msq0+3uDABOp5bgSjIzjCtoL+HrlrrdZDrado4bv9U95McA8E79l9FMZOecDIKH98S1+vVBGjrg3Lu/9It7wz/4relwaXozVLlZJkdK4a/88lg4sv/OOsq8XgdmPXe8FAHmCu5B3XqCMjPG3jtXi87DDPtcxB1mieoyEHiQElAz277+6nqYmYNiofXJC104D2vhr/+7E+HYFO0y00gKJAWSAkmBpEBS4KFWIAGsh3p70+KSAkmBpEBS4E4qsEZJ2zfeWAX6kJekewmIIbQoFbYo16Mr3YDJ4yFmJlnSZ8c1DUICqIMTuXCJcr8nj9z8RtsudxuVBm/QOaMAh390+eRx/7x3YYOSOa/oo5arWVbVhTsLh9RFusABl5pYfzKU4OkWogkuoCbgXGrnOzUgQrpccjigdIVd4PUvv7UWLl4ltJvnonuI6/5zyiL7fnaMLmzNUAJirAHFpilXPH0ZeLUKZIiOI84LBMt2l8N7Z/MAjgJr7wrLnL+XtRkIPg+Y0gU23pPB6SXca4YloNnsYj1c4Xw5AJf5W5NkJ2WBf+OEfR+fyscyMZYXoaC6Ciy6ukqxs93aBqWSwKMjlE46rx7oUx/XE76M9JXDBQLsNxvAM8sQAYkciRNqC+dXM659ZICA8aV6eP0ka8uvEyzOuYAzxw7QXREn1g8a6u7QmXQIx5IdEYUpOsh0hKn72CDlcQ8YwHJN5oCdoMT0P//1AzgL1wGza+FzL0xdz63yNXdyuGcC3Sofz9L4YGYxA7Rthit0rTxLKL5dH20QsIHG3rPjwFU/CplaIKkr3u/Op+10w6W4uhn+6VcXwt/8iwfu5DTTuZICSYGkQFIgKZAU2IUKJIC1CzclTSkpkBRICiQFdp8Cupj+r3+5EC4BfnSHlAhGF+bIlob78+HP/tR4LI8Sckwv1MMMeVJCJ0eNTKjz09VYKoc/Kj62/ZNB17q5dD51RgvLjxlXq+s+hhMLWGLmlJDm9IUKAK0RywE9ZgXAI8TqjiVXZAYBVHpxHekaWhX48G5/mU6BlufpYprFeeU1YzoVzqnQomSL8PczVyphhOnBniJUuECnxU3yqjrwyplsYW9aWWuFd8+XyU/CHSaMYH2CN8v9BCIYxDi+RVe6Wrg6w/UIaF9epYsfc7HMsauLMseeFt39LIMUsNWjs0ntBFOCQh02ozwvYMplG6wFSMfz48OUF3odANcY3QFHgWUGqr9/vhLPo8tnhFI0wZhuH+c8A0j0NWZ3VXCYbVBuOUeJ2gqw5GnC1weAKh81hCWeN2Z14bh6/FCRzKt6nIdz3Duej+H6H3WO3fxcp+lAX6EYFhbKdw1edTTQxeaHHQpHceBdnK5HwMgtjWOwES5xf9owAMlx/jXYc37XeO7ybIX7zB31N8tCw/aYpUOhvwednLdrD6cvSYGkQFIgKZAUSAo8ZAokgPWQbWhaTlIgKZAUSArcHQX+19+ZDX/w6nLM4sHsgzvK8jfzmzLhCdxEh6+VW+kwWivzbvuG4Ztzu+rtNCwZzBNULaDpQCwzr2ZwLW3izBKeVKyt46sh7r55t3PfJmWMlh4abG3IeXGLjCf+za47aIw3/ZYcOuBZEVhdpERsmnOurNF9kGNDV9uBpXEoV+8GOuEk622XeQnUzNhq8mQ0gPFyM7B8rZytqdtqqRlLCrt47TprNqOrxZOCKDv3Lei6EpYxH3OzXJtdALt4rV3mtlhTkVywLIRIWCe88vxCK+HbOueTjBmergtszXVSa3ZwLx0XCfG2I6OKmgcmCLlMx0A76A3beREoxTTi0M3jdeucR+2Ebp7fssxLs9Vw/EDhugOsfcTNn4VchsrrWitSqqlrbJBA+k6+2M1HpEduRwED2t3HPsoFLcsVNtbJTHN/qCzEkdUMZwC2L5DHVcj6ZytgmPvNrfU+86MfmJzg1e2onV6TFEgKJAWSAkmBB1uBBLAe7P1Ls08KJAWSAkmBe6DA99/bCG+TGwX3wNXTCTEnYL2ZCf1AjKeP9l6fhbDHN9S9gBnLAjtDCDVJSPlOo4AbxRD2CVxDurcENkIygZC5U4IkS6yuzOJoAgoJbYrEE/VxHUPJuwBR3bzb99qCnCIuoZO86beDXj+gx3LHGp3zFgiVn8cNFTkAEEAfi64pRy5LhhXPTY3qfGnP1Xyqq7PdoSGYuvYXg1BIh9dAbz7CA1FCLOficedleeMqrhlRkeDIzoSbrK0BWDJMvQFUc112WBSETU3kY/i5fM4ZCbp0BDk9j72E62aNcsmNTZ1m7c6JdcR5jowsoZ8A0YD4PGsWZOmSGmfegi+1MuOrp0uHmjoB89jED6ATc+B6OsjImv/I4XG6rXSuCeyEYGncvgKx5JY98d65MRi+SBOEItllqzgEa9z7KhvvJU/PN8u4+QaBm1/4VF9Y/kPKW8nNcj/8nSqx71/6JDWzaSQFkgJJgaRAUiAp8NArkADWQ7/FaYFJgaRAUiAp8HEVOENQuLHdvaWuUKazXh4gomvIkPIDk4XwxNF2WWCWN+d9AKOxoVzYO9GkZK87uqAEPJa0TfHanUYvb95jtz/euR+gI5wlcrqq7Grom/R3CWo/c6lMptMHRxuKXrYkj3+TDxLYLgQTJpVxN3XDpK7i1BI2Te0r4MBql/h18TULPZikM2EVoFWpcQRQgWmHkSFLDrfIQWqE44fpEJfPhx97cQBwhiOmWY3rEADl7SZICPvx/bnomBrs646leOKnjUwTQIGLDOiwuFilTK8RywYrgCidMnmywFqIoVumr5ihFK8U9o9nI6DbM8r5gFH4reIihVzvnC1TGrgRVtGincPeFbO2SkCOy8C8/RPCOjLGcFxhCIuwT0dPi+cFWcJFwdWVOULmOes6DirhnQClxQPOQ1DYkhLe5vD8njuN21NArVYAUx3Nurqa8b7phLl7FmHvoclsLMuNe8P+sDUROObJNPO+6cWl9cLj/eHJQ/nwL/5oPToJ+/m9+dInBrlPE8C6vd1Ir0oKJAWSAkmBpMCDrUACWA/2/qXZJwWSAg+xAl/7zkp47f2NoEHmMcrTfuknRx/i1e7upWV62mVjjVZP7IxnuZwlgfvGc+FP/+QIAeOWkhnm3s5SOnogH6GM8MnsJl0nTxzpo1SKd+a3GJbP+SG4efcsHQbJo5oGYF2dI0OK7oTb4dX2U5jNZY7UUD+IjWttAqXyOV1CZFWR8VTHiTWB80sIYJZVAQA1TGe3fLZEThf5VgCevQCtvcAgocI6Dd50PjU5l06lr7w0GN45VyaAne5wXGcP53rxRC/5VbkwOkCZHkvOcT3Lv8bQZx2n1CrB7RdndF41cG8BieA9LVxgQgy7BBZ0j/V3R4jm9YUW5lJZ/mcnuhqPnaaD4Gk6ORokbwC7JYCGvlsa2GqWQl++Hsa4fgnNLEHTFaVzTBhocL1w0bU6Thxq4UjLEYi/2Q6kB1oZCm8m1pVmLeYx5TM6yrYre/P3ll1u4iBjmhHIxbwvdE3j1gqscy904JWvUjsfE3QKNTvjwJ58+PTTfWEax10FuBoZIfvp6x472Mt9OBz3DJxFuW4v923bTZeccB0F09ekQFIgKZAUSAo8/AokgPXw73FaYVIgKfAAKvDbX1+iXfzy9ZkvLFWj4+Cv/uk91x9L39w7BZ55rIgTCMsTw3K0EgQmj+Pp1/7EeJjac7OrSvDz7PFieGyqQPg05U84iwRKP2jownr1/TKB57VrpYGU4wF0VghA/6ghFCDDPXbnY3qxhFFXVZ0H18rtEsN+XEqjgKsyeVTdlByW8s6JnCvAEYcDcwBJFP4Znv27f7QcLs5uRpBVBGrtAdSZ8WXulCV/wp8sTqR1gI5usRqxRMIIg+1HhrrCd+ly2IAwyCcyOXK2mFN74FzjL4/eQhaXlI4o50dI90i+7UDjRXYrvIS7yi6OCwSt6yRb36zHTCShoeu6SDD8mels+CVytA7iFFtcNdmKMjRKD4eBWoUbSJRzfYqOdycOF8M5Ot3pqDs3XUZXujMCoK4CCpfXeikFvXUdoXli5nR1hm4190YYKKh8WIdZbEJG3WoflF7e/moFrDcO79c658xv003g+By/ZwtLA+H7pzZpkKBbjlLWsVz4i78weg1etc+kCy7ceqtuvFz6OSmQFEgKJAWSAkmBh0SBBLAeko1My0gKJAUeLgW+9cbqTQs6d2UzXJpvhQNjP8AmctOR6YGPq8DR/cXwlc8OhG++UQ4bABeDx3/sOcuZSrc8tXCoSJe+4vVeabd86fUnlii1El45Cvku3rRncTfVImC6/qIdvrF00c6B/QRhDw7QgRCQowupCJzpx7E0gCtJaJWjDq+PudsV0TUsk1WVBVDpYBIWLJEtdO4KJYP1FcASzhn+yWOxWiGDapISv2eO9ZJVRMg7c6MoMbqihjg2P4JzDJBlWZ45VeZRGeYu9CgwtyYQy2t65/b09HB+juV5XWstUuLzXNs8K11cgiI7/M3MV2OZ4CIh3jUhCOzI43OUlHV11cKVma3wL7+5Gn7xS0MAsFycj8eateT8dhpcOkzy2nOXKjEA33JIxyaw5M3TG6xx4JZcxAB+IaGvjYH8OOV0hQngJgFwdtW7HUi507x262M6pXScdYbZan1Awx9qcN/tdAPrlts+BIITo/nwEy8NhaOE6l9i/wu0szyOm7HEfR0B7Q956e3nT98nBZICSYGkQFIgKfDgK5AA1oO/h2kFSYGkwEOmQKUCQMC1s9NYXCKYe+zW0GSnY9Jjd0aBF070Bz+EJOZN3Y1RA4p0hm4Xw9w1m5SABh/lwurL46gCRI0QMq4zrMwco0sKeGReVRfp7i1sLw0g0z4cVMf2F0IX4Vl7xwthboGSLa7rtc5dNeRdUNOg3LBdKtfskUp147hqhPPT1QjlLHWs1XF9QRUGewuxxHFhpREdSnmumcv2ANTwRHGokKi+RcYXwEqHGU9H2GappC6sUboldrNWXTUCNB1iLZ1jhLyvAFAaXKcDQFTH0rF23Dvh3rz+5IVqGCa03Wwv5+VaPmp/nMMFcr2WAGNeR+eYJZ5mYc1QJrlvxPN/eOg0m8cN5hrn6JL3vXfXI+hrYi+aHC2GzwAzX8Q9dAhdHxY3lsBOeCVY0jEljBTembm2Pb/qw0rd/JNltYKw7cN7+8bSv+jg496dGDbPjLLW8XwoCIC5r21OYPmmnSDTSAokBZICSYGkQFLg0VUgAaxHd+/TypMCSYFdqkChkKObWja+md8+Rd8YH917c7lSZe6iAABAAElEQVTa9tek7+++Ah8FRz7u1YcoR8tBCGqAkVlAiSBzkJK4Q3ty4UytQj6VxX4fHuZQ9dF9T2eMYGAfeVaGlQ/094SDE4V23hXOq4WVajtDCsjjsJNfH06lPtxlgonphVoo4NiKhhnyr4RGMCfAU/u1TToueswqZX/mTAkUBEl2CDxNyLxlfoKzKnPMAq16uQ7YKlQ4oW6wPCVi5mUVcIc5zxwfzntsiBI8nnNYbraMC03YtomzyWD2aNTZtux4Vc7ZQyB94MMyyR4Ihy4yIYuw5aPGNJDq7OVqWFyuxrJEL6CzyHnMAqn2DH34BGtoKRSzi+NlAN4fvroSSxz92Qq45lYlfPsN8sKYiyDNTogPw1hjn7/+ykqYX6IpADDyAODzueOl2IXxhwFYOtMcHRhmyaaNDnYaugUt0xS+OvRodV5rKWMaSYGkQFIgKZAUSAo82go8HH9lPdp7mFafFEgKPIQK/Nznh8Nv/N4ibxbbATI91Id98RMDoe8jQsAfQhkeuSUJBo7sz4ZTF5oxDF14s4fuhU8cLIWDezfpyrcRZuYox8OGVMCxFLsPAoQGeePvm/9x3ExHAFLjI0AwKhH7gVq6koQGupXMn3KYFyXwqVKbF0vyIEuWEgobLOvrws0E3gnd3H5CmhyfxgE8E8PmYAGi+HkIUKM75yJupjXglSWCFeCDmKFRJyie50eAaEKpKqV2FSCZQMKPHi4yMkgAOw4bM7oO7iWE3eM4XwmY5Kt6cEXlr+Uc2VXx2q9CgKXgBsvgzKE8kRKzPPPWzSO8cj4Yh2JppMC3BCi5xsY4ZwBcVcL3yRhTB8sYK3zEzojoY0nbEL9fS+RpHY6vBt5xLp1Ijk3WtrxRj2C5hpPN53zGDLCFlTo5XDXcWWRi7WKAZXmlLrIqLrchylP76F4pTLxxuDabSJwiRN8MNwHhEuuDFYaf+BE6/nlfdUDWjdfa/rP7KEhsK9sOae842rwP00gKJAWSAkmBpEBS4NFWIAGsR3v/0+qTAkmBXarAc3R527e3N7z82hJunFb4xON9lCfRui2Nh14BAdTIQA644lJxYNHhUDizZzwbnjxcivlWa5uN8Oo7GzFEXfeQb/iFUsKqEq6qyeEcpV5tGCXUETwcxL1XpiugriazoIRQAi878a3X6zFfStfUGCHslVo2BpQLsQxGP7ivQAB6KYzxeuczhtNIt9UaeVfLdDqkyIsSPEAH1yySUeU1+npzsWNhUeC11UVwepljzLyijFDnF+fN4wyzDLHE94KKfWPc47xWVnH2Eo4u1mnJGksATuHu4ngdcIPM01D6A5NZju8Kb58px7XvIfB7jDJK3T6CGYGNrjYh1hydHM8AsBZXapRbmk9GuWEVkSlvLDCPo5T/CZ/W17aiC8h8K0sLLTl0tJhXNyAnwzxjRrzzvAZVLD/kUT7tXsqyQsnk995ZJ1eM7pZkjIVWd3j8UDF88aUBQCduOa1318YV8qfeO7cR5lercS9yAMdqvRugVQ5f/GR/52V35av7632s/p2hrDYQSCMpkBRICiQFkgJJgUdbgQSwHu39T6tPCiQFdrECYwMh/PEvDO/iGaap3S0FBnEuHSPIemU9Uqx4GR0sQhah06W5OmChK5y+WKWEzk6AIXbf20du0AAdDwVag7iJdHRd4y/RJZTNZHES0XUQNpDB7YKBKewF+mxUM2F4Lhv2DPeEYk8llOsFOgA2ApVzYT8dCF94vDfCK68j0BLgFIFQzUZPGB0U3pChBcQRYumi6QLyHNqbD/sncsyhi9fkwgTXeefMRlyL5WBWKe4lGH4Ch5kB7kI287KOAGqHWL/f/6uXW2GWLoRduoT43xCOoaP7+nAitt1S9Qad8Sjd6wK+GDQfgRoZSrq+yji+9hDYnsPFZYniDO4og+kNmzfTyjI1SxqFVFPAvam9OeawhZOqFS5MV+I8xVHmc9lFUreYrrViMRMKZUoKnTBkRfAmXBkZZL1Axt04XJeQ7yIlkKcubMYAfef57beqZHvVwk9/diiuodsWloyXaVbwynsbocENkGfdQ/2URwIC65RNtraBpfjiu/BJnTfJQNPppr7RGch9lUZSICmQFEgKJAWSAo+2AglgPdr7n1afFEgKJAWSArtUgaN0X3vvXAs3kKlPlM3xBv44UEtuMtRLsDZwSIi1SlaUb+0FMoZcW+I3MpiN8MrjOqagtoOFgHPAUHycT8YKwQdCP1CmtL8nXJpphcP7ydMaGIwdDMcoRTSPTbeWkKi8GYAKzofSO8DPOOWOVdxO718EJAGJ7GpoXd0IsGMMkJRjziMD2Rgub0fGJvBD949un3FcYkN0TOyGhpl75Gu9jo4xYcmnnuoPU3sK4c2TG2GR8w6T9XV4X4njDKTvDrNkdpkTJoBSG8P1vf4ZOgw2oWPq9BqLP3GlGJ5/og9HTytq4TyFd1vMRcCX5efRwfbX9XI9Oq5WcSsJfQRf6lWgi+IQOowPN8Kzj/WGd4Bmi6sEzHPOfhxpn31mIHz66d6ofRR3l33SZbeKW25+qR7hVYO1uz67O86g4VnA1uNThbBEKeQr726G3/zaegRb7mWPex67VuKuw16Xu3b/3M0lNltmmZlzBmRlj9yvNJICSYGkQFIgKZAUSAokgJXugaRAUiApkBRICuxCBYQrz+N80lkkjBFOWRZ3abYWpoEOwgbBje4godVeyu90RenC2ilvyMeyHG+mk1BLh9MK5X+6loRQupEmAFZb9Z7w2LG+WGJnHpYQYfuQX5kfZRmiDi6tUUUcSBdmyIACTjUBD+YY9TOnPuY2eC0TylLHZx8rhUP7BD/CE4FYk7mSM4XTp0pml50It7iesMnvDQ7fQyi9MxBueZwd7QRHLkKwkWfNDt06q2g1jy69rHUdaLPC2s7RpfDifCV85pnBUAfE9ALrwH/RiZYDik3iAjuI+yp2bySA3qpCr6/ma2RebVEqWOKYkf7u2MGxr5QJhyZzdDFssM4cUFGnWT7uRZzILvzkHnrv6D6zNE+tLE3WBSc8nJ6thr3svfv61W/RLKAMQOI5NYBzUSbaCHuqWUqZi7Hs824uMc4HOOlcdPIJOy2pHUT/G+/FuzmPdO6kQFIgKZAUSAokBXafAglg7b49STNKCiQFkgJJgaTAdQXMnOqMizPV8PbpzXB1oRrBg5Dr8L4i8CRQ9laNbqFLhKrvB2Y9caQYQVXnWL8KfDpulgiRCPO+ArwQ+ggz+kt8UAVnQPutRhZwJQjbPszE8qOJpasCJBJOCbeEZoKnzhCi6M4SoizQ8c9Swa6ORYwX+fgIGVYU5sW52wFwq9Y+3uf8sMOh+VftjCtK2iQsjCzAS0Bjz7sFOhmev1IBvHA8z59DmzJh8S+e6Kdcssa8emKW1xFKB8cpYRSWGXS/uNKGaUscLzgUojiRbhxXFUogzWby48mjpeje8piP0ipObBd80v02Dmwz62qWLDDdV5aRWrpp50l5YMxCAw7OLQPxrB1lRGX5Vrg5NlIITx/vvemeupPL03V1lfvXzpLu6+xCG65eyXNPTxZih01LWNNICiQFkgJJgaRAUuDRVODWf6E+mnqkVScFkgJJgaRAUmBXKiC8eZ0OenYTFOnocLIEcI6yMN/s5/k3+hZASzB1fnoTUBXCYweLt1zLBs6bOSDSmSsArEWztABcgJrDE63wLI4sw7R/2GHYfG+Rjx9woPDJ1944BBhCMJ1ljgjbcGZVACobuLU6wzB1XU8VnptfqUZnzgiliKOUTiJFOH25TLfAtnNNPlZrNsI3vr8WXWWH9uQ4P2WSdGjsidcxs6tdommZpUBtFk3NChPACbFOXqiEN06WyewC1OHYEsIJFu80vLLToQDQ6+qQ28b2Okv/kb66phOHC4DFgTDD2tZxpunaU4MRwJbOO69bphzV/DDhloMKyoABKrrQPnGiFF2A7Wfu/GcD9xdXG/F+dj/KuuEArI4KYNJA/3NXWmE/+5fj3vlR7s87P+t0xqRAUiApkBRICiQF7qUCCWDdS7XTtZICSYGkQFIgKfAjKnD+aiVs0EVQ90z84Dw6aDaADiU66sXOeNvOfZVyvuNTN7uwOi+xTM68qKVVMqkAAjYzFH6dvfr/s/fm0ZFmZ5nnK4UiQvuSknLfasnK2hdX2eXybtr7RjNAA6fBdIMxNDR9Du4ezMz80cAwvUDTMD4DngaaA92mwTCNsaEAY3CV96XKW+17VuWuzNS+hBQRkuZ5rjJUSlVmZSqlzFTo+11bGdu33Pu7X0gRTz3vc6sqUZyJvdvOLX7VjnGxtxZmLGBZrKq1skSqaTue1IoS4iywOH+qNN2QSg1r27mMzM/b5XW7RJXBsbxK3iS4icVOuXTu/eKQHkuFUfN5LMO5DHCqVImvPzwusa8lrt7WnrKVtstVZXGnlgvWoxyxrraGGB+yaqNcLfVnRFlXYxLDhiUcHjwhB5b6tbW3GLslpNwqR5LLJVfb7H775P3D8dDTEymrqlPlct91V08Kz+9Q6ehaNLvM7ryhPYlQTzxfSvyaNQddKqN01pdXrZwVra4OsZUTraqxJxlRAFs15mtUKmlhy0zXujnU3xldvv58Hfv6HlK4fLPKQ32dWNi0oDU3X4nW0ca0aqVXmlzu4FvrfnE8CEAAAhCAAATWF4HVf+paX+OhNxCAAAQgAIENR8Bf8J1JVFU+k+/PyC3jskGLOHYjNcg5lZMAsbxZCLDb6WzNYsSkhB+LA0tbVY9Hx/XctqXPrv39drmdLFr49HbfjMthZSHOzSVsHmePShmd62XxxUKTHUEWtyyi2HVm4adXIfG9WrHTzSsa5nK9WkVvXMdQHpc30LZG4DynWf0zpQUGp5RePjKuUkm5qHyOWrPryVlL5Ya8tptV3thMcn6NKQtrUplac9q/qpUXHRxvoWXn5maVtr3oGKsdZ6W3935hML72yPhC5pZ2dh8/dd9QCubfv6cllSuu9Ji17Z155mywUZVTWrQsSqnct6uY8qW8jRcCcCmqV76s6gLbu018BXpUZZRS8ZQn1hS3X98m1g0SUGcl4K2NoFbrn2/tsPN1YNegSwSdcyb9MF3fLZq8vALX5vwGUKsJaL4+LGpVtPzluK5j1zt6IQOvYEiDAAQgAAEIQGBjEnjxU9vGHB+jggAEIAABCNQ9AZcPnpITaHCsrGByuY38RV/B4oo2ipuvbk6ZT40NC1/wa4P1qn3nEq+8TYeEiGYLRmO1PeyKkmNJQoBFo0vdavlRFi/KEiE6lc/kHKtasxvn5LBzqJzLtFDX1iFxwo6cWomhBRkLWz6WBRCLG/t2F+PNr+yM+x8YiamZhUB4HT6aJNL0SOyy8HVS7h67vzoUyL6jv5AElDHlbTkc3sfaJSfX80dLEr50QB3Xq+I5yqvqW23jPue1nUvb7ATKO+3+IpsFxKcPSiCTu856W1X981iGJ8rxjccnYpdKJS0wXUyzyDOs68ZOO4f1m1VeGWfOEOvVCpBmWVSomcs+ne21c3Nerj0tAqDVBt0fdSNadW6XHzos33NyMc3lnxYpjbNZAlpRfSip/NMDdnh+OpEe2l1nQdG3kxKxXCLq68IrEXqu8xIOXV5ZawNyGfp5h/K7uc/VaiF618AVVzvHSm8nlbX2/PFppbjNxtb+5jUvM11pf9geAhCAAAQgsJEIIGBtpNlkLBCAAAQgsCEJOKD9qQMSVCQ4NOnL/YScMEWt3nfDNc3xqps7YkCrEj7+XEkh5ioCa8jF5t6muG63EtlfpjXL6XLTNa1yr8xKhPF+Enl0/ObTWVIvs+uaveTz2dHjnCmLLUubRY+KnhuSePXlh8bj8RdK0SAB6ZodzfG213TKbVWQg2s2OaQs8FmAcTi7c50+8I+3xPR0Y3z9kaG0qqDtUg5r397vwHZnS6l8UUKRnWwWi9wKysMq67HFrmtVHlgWS4s/7keuSRljFsnMSMeyeOUcJh/Lgthqmssbyxqnj1PxsWqH062dSC6HdLD9xTRnah07OS1Osu6p+VgeY0nOpbaiQtElWNlN5hUJ7dAygzv2acXF7pbkPnMGmTnvUdi9V8FcGsh/of1xkL4D+xfEL4tYC0KYg+HdioWcxLQXx2eHXZ+ETM+nHYbm4TlI17WyumrNjrqxqcpLxD2fy/NiYdHXVbeO7XLUS91GdJ3+P/9zIB59djJlyTmjzSWmd9/SEdfsar7Up+f4EIAABCAAgUwQePGTQCaGyyAhAAEIQAAC9UfgmUNydEhEaZ7Xank9yiqqKLNIX/T3bm1OrpatChb3KoB22VgUapGz6uXcVzUCqZRMJWTPHppKzh87cToLZQkmF+8oqh17JbcuEazOLgR21/ablgDhssIvKHz9wUfG5MaZTaLTieHpODhQig98z1Y5iXIpbN0Cll1R/T0L+Umb5D77uR/ZEt96siUefUY2HrnTTsnNlcs510llgtrPolCvthuTSGRxyqHgFkkqytyyU+iqHa2pLNN6UqNEK6/WaFGkKIGsTXz941X9LIB5tcQkorVpfs5Sylkb09ludejY2Z+LYwMNMSO1xs4we+mSG6ndpYoXJr7YveWyRwuB3teleDWHmM87NDobJyR0DirzzC4hl5xulSB0/d7W2L+nObmcxiV0zekau/vWrtg5MJ0cYW0ao7Om7HxqXcEKgBbPhiQmPSsnm9m2Sly0IHZEK2k6pH2b8sfcHNQ+NDafrt9aKaGFMgfyW3iyc8uln5MS3ZaWu1rwOxvrwdFqfPvJyfTja8YC3Ntes5AndoEoU79W8o/LS3/l9w/LSTela9Szp2tC15WdYS16T22RoNwutx8NAhCAAAQgAIHVEeCv6er4sTcEIAABCEDgkhKwoFGVLciukpxq/Py4QVlNFgNc2lZrFqxcIreSZqHguj3FuHpHIQkfs1qtb/DUeHLJrOQ4q93WwkhZ5XhLXVgOlrdb6PHnppKzrOZ0sphhx9k3nxyPa3e2iMmC2DMjk5EFjc52lw3qvlDcsb9TgeszcWigLJGqFIOjcyknqVMi1BYJG309BYk0Fhxe5OjjHJLjzcHm8lzF3u1FrdpXiU7nccnU1pzmQa+of8dPqczvsQmt+ijpSUqWRcQ337VJuVhNMSzByH1zKafD9l+uffcb+7Q6YDUeP2ABRIKaxMnrdjfH7u2tyR31cvv6tVRyqJD509pJVNRvCz8ugXRumFdkfPJgKY5LhBuSq0wRYApvl8A10xQTEl8mJBTuU+C/V6Gc0TwU9Onwpmtblfm1IAj6OvEcnUsUHZJo5GPY1ebVGXX6OCI2B1S6aHHUitVOudp2b5GzTWLji7QXRuZ5tojpDDQfw0KTRTg3X/cKwUqOw6l0Dr9mrs7KcjnnQmmktx+UC+qvvzgS33l6XNeOXFt6ruVwKc3Ph/7pzrhW5aUWJ2vNc2ZH2EqaRdVRMZySQOfyRYtzX310TAsfWPBbEK98PDvOjgxUYvRqLQKgst/2lzdErqQLbAsBCEAAAhDILAEErMxOPQOHAAQgAIF6IOCywc62Qsq/sjKw8MXeOVC56LvI0rLl47YY5p9KZWVf5pcf52IfW0PoUamXRRcLVe6LbuR6KksoWAh6txPKzaKEtxkcqqokcE7OHmU5SdBwz+14sQtpwUG2MJZtfcW0UuDOLcUYHHYY+4KA4/LD4fFJCSZyLkkI8XkcZTWgkrv+/lLKXrIwNDGj8kRlRnXqx2WXFhDdh6JUluePTMfDz5RUWijRR/Pk8PmPf3og7tBqf3aBuR09odB0OZzs1jpXa9fKh+9/95b4zNcHY3zCq+vlUzZXX48yv3Quu7tquV9nO4adSzawnRwuS0iR2Ckhql+uPDvLGiQeHTpeiWGNfUIuJrvITGZG46pINJrQ6oxHT6hUsMeOtAUxzOztgLMgd752TELVUHJ1qQ/a8YjEQgfHD0vU+o5WVTw1Wk7ns+hYmWlNzjeLk8ubRauFn5e+5m2FIZUV1var5ZANy73lDDX396sPTSTxyo44X1M5cZgTmKMnpuMvv3gyfvidW9I8zGlipzV+C6ae8y6Jk8vFOWtRFqQW+tQop5+O/8ikVumcSmPL66K7VqWBfg8eEV/Xl7rntevU92Y0byFHW22Vy1rfuYUABCAAAQhA4OIIIGBdHDf2ggAEIAABCFw2AvuvaolHnnaZnIO+Fxwq++TQcZnYRmrJbZNkAAlaKlvbrDBuB7eXJEpZHGiQimFBx04ri1RLq/Xs4HGz+LK8tEx7id9EfPXhsbQin5clbLPrS84fl5k5E6pVmWLDWnmvWi3HjpHJlJ00a1eYXG52Ubnc0I4sr/5oYasit9pBOZrsmLKLZ15CyIxElJEZPX9sRgLWguXGuWQvHJ2Jm+VoOlfzanrqedymzCSXo1kI8nidt2Xnk8/3cu2UxKm//epIfPGbo0mgsivIKyS+5VVdsWNzQcKPws+VNZbT+MzYYpAFFxmIrLMkgSzd10O/ZkHQ5zyfQ8nC2pgyrtx/99nNpaCHJWI9c3gmZW85h8riWmmmHM8qCH5Td1cq/Uwbn/6nTeV1ywWkpa8vv+/3gPPB2lViWFDW1JFBlUa6PFKimcUs9yT1X+f1vHhspyTs+Tlv0yhha7I0Hw6B9/x0tObjtuta0hgsvt334Eh86dsTqSS3t2thUYBpzavLH6ddpqkD+ucZubtaiy3pGvV1mZdzziJmrbVqlYW92wu6Vs8vBNb24RYCEIAABCAAgXMTQMA6NxtegQAEIAABCKwLAtv78vqi3J5K2/zlfHNPPrbquY3cLEhdpbK2d7x6U/zF50/JJbUgQBRUPnmtnr9+r8oHpWB5ZTq7ZCz02BnVLcHL7ptas9jx8b8bjC99aySF34/J4TQr8cHh3h1yPrkMzcLGlELfp+XCKkuAOH6yLMFlNpULVqo+mErFJPw4x8huGm+vuCyJibWzeFbklNK5ZvWiS8uWtikdy9u6n2drte1bLTJJXEtldBqEVwz0WF7OfTU0Vo0HnpiMT94/GKWSagPVbHAanaimkrx//Kbu2CSHUUkOK5dWNowqy0uCk8XAFgksXoXQK1o6F8xZYeboZkHpfOV1Drx35pPFnFpzmeu0869GXdI4lxxOfs2EhsV+q4L0+1XqWkpCj0oTlRHlLK6VNAtnngM3Xw8uCX3e5bTqv52JFjxrzZt5LL1ySpn/tOZiUq8/f3Q6hfQPSPwaVybYP3y9MV5za0s8e3A6HnyilI7l/Y6dmo3/+fczsUdlts49q53Xx3d+15j2vWFPaxyTW/DoSYl2EvRmdB0VWwvxfW/ZFK+7raPWFW4hAAEIQAACEFglgZV9YljlydgdAhCAAAQgAIGLI+Ag7e6Oc7t4Lu6o63sviynf/aae5CL6vAQou2Z2bS3GXTe0Khi7eVEccRmg3UIWelIG05JhHTpejm8/NRHTUltcMmZRpaptByX8NDXlJUJYMJLQ0jSX3Fuu+nI5oS1QTXJdpVUHJZL5HBZq8iolbJRFqVcZTMd7qjGqfKnainoy4Ng2pRUPFwLKa91oOi2y1R4vv10QirQyoDOg1Mdaa9J5PJ6lglztNd9aTDmo8X3todEkvtXEFY+hpFLJA8dKGm9P6o+dXSmUS2LcUIMC7dWnLo3lauWIXX9VswTRYszPlUPxTnJhNbzEJbX0vLX7LlE0HxV1SqBa6HcqyWvPJxfXnEQli0ZFiWUWg7ySoUvu7AqrucwsOq60Ld/H/XVZoLPDtkrctUtKxqqkmuV13m39hXjdHaqPXHhKYfbVlAt2QM64wZFycr15wo+emomynGKzsm0l5vrHrrtJiV4Hj83GTVc3KY7Lwt4Cex+vscG5X7n4wbf1xgvHynFIrrxOlYu++7WbokPvWRoEIAABCEAAAmtHAAFr7VhyJAhAAAIQgAAE1piAxZ9X39oRd93Ukcr9LFbYpWTxxM2ZTs7Osg7ilfKcn7W0nVAu1IxKL5PoIbGhQT9WNuzMckbRvB5bTNEij0kEW9jXStScXs9ptbxcEncKOq7Fi93biinfymHvd90k15bKEE+NzCQho7u9IKdPMXZtPtMdt733zMdL++f77rddRe6XBau0kqDG2SfX1HKxpravt7F7bGxCuV7Kc7KoUht5Tcgq6/licqypPlKHH5uqpsD+GTmkWlU2t1vB6nv04/B/H29MYlx1xiWTC26l2rnOdWtxql99HJNY5lJBN/f3jhvaJAbJkaSyvqqUJM9hi/qxZ1uzHHILHz0XstzOdeSXf95CpZ13i841rwipn50SxiyU2Q82qBB987hOTr33vqFHQl1zOqgD9V0maEHPoe9mWNHYc7oOpnV/SGWWnVr90dfYvIRNpWCp/3LBSZO0K9DbunTXLjYfy+w6tfqkr4sbr2l7+Y7zKgQgAAEIQAACqyKAgLUqfOwMAQhAAAIQgMDlIGBnT7tK7JY3iz8t0mfO1TZ1FSQ85JJAVJCjqawDObdKUktsklNoStlGUxJi7KQpJDfRbPTIJVTIKQhdbp7dEqMsaDgUfcumvDKNmlMOlwWwXQqGv/GqVpV2liRsyJWlXK4+5XYdP2WHTyWJKf3axyWfL9ecS9UhEWTqdP6VV+NrU77TUvHKoeXSU5LrzLdePdDlcJaNupwFpWNMq09JvEpilgQwjcNiT5tKBO+6Mb/gMJOVrHb8pX2y8GPHW0lCk4WpC209yne6sZhLK0NahGvXMVwK+fbX9sTXHh6NEZXUzau/vQqEv/P6trRq34Ue++W2s2g0qUo/r2roMsSrFag+rcyylmJTbO/znDfGPolXN+xtXcz0smi2pbcpehSs/+izCn5XCaB5WeialwJoJ12zcrpcEekVJFPTi3k9uOemdq3cmE+B8E06zrbeXNy6T05Aza9D4GkQgAAEIAABCFx6AvzFvfSMOQMEIAABCEAAAleIwBblK92+ry2++uh4FIsOXM9FUSLL9r7muGZnQdlWuRhWvpZXH7QLJycho7+3Jbl0NqsUcLtCwvu6LSgpyF1iWM05ZHHJJYMWjq6/6kznze5tBTlyziwjPN/w7ebxz/JmoWxErqBaSLoFF6+I1yrRyi6kXrml9ktEO6rcruPKc7KI49JDC24/8M4+lZ1KsNE2dmgttJoyU3u8+lsLX1dJ2Ks5v3yuzrYmMWuIQyqrs7urQ+WDFnp832LZapv5+5i1ZibX7WmJEdVATiWHWWP0SXDq0Tk9R0moOn3aWxWo/8RzU/F8vizlajYak2Alnvn52K3tK1GIcQmQLu30aoM3Klz/bfd0J0Fxpuwy3oYkdNXOzS0EIAABCEAAApeHAALW5eHMWSAAAQhAAAIQuAIE2pVH9K43bIo+OW+eOlCS06YxLGDccHWLyt4WHDfDysM6ohDu0fGZmJgci2v2dCQBJpX0dTjPyS6mxrTq3eUegsWYmnjlc/t+Sc6hnJxVFa+SKEHIpXN339wR33xyUivnzUVba0Pcub8jBdo7G+pF8erS9n7peezi6pM7S5qgShznEkMLV1490K6wmhC4Fj2qyoXl7LBRCX2ey2PKoTqiEkYHur/prq6457b2JJpZxBpWmaTLTm/b365VEufj+WPTEqZUDqhPxL7d3T8fd9+2WYH/c3LEzcY1cnZdq59a8wqPNAhAAAIQgAAErgwBBKwrw52zQgACEIAABCBwmQhskgPp7VrN8M13KitL+oMFFLdRlbeV7VjSY5eW5RXk7iwlO5bsYtqmUjQLV1eyeaW/pc0ikUWrw8e1ip4EmykFlnuVwxNDswprV0B6v0oPtY1LIx9+aireIAHnSrUxiVWnlDNVPr1K4fBYRW62QioFzalUby2aw/VdpjgwXJEgVYmnD5fjoacV2q8VGf3ak89Pxv3fbkvB/+Pjdn/NpzD5DjnE3npPZzx3uKgVExXoL/GqRznvO3qrcbNKA52BtVSQW4u+cgwIQAACEIAABFZHAAFrdfzYGwIQgAAEIACB8xBwNtKpEWVCSQvqlYCxkoyl8xz6gl+2GOFA86XNq8dNK7TczqBx/bS3aHXCwcYUNF5QfaBL4y602d1jt9NaOot87oXQ+Rd74eNX5RzzKorOwBq3SDRc1WqEpShLsfEKeM0qd5yQe2ha3O+6aWH1vRePcHnuufRxWMJQTbzyWZMDSiLWZpU3vuhpWl1/7LganajGuH5cavnUgQm50CrpXD7fCQlbR78yHA8+MqZ8rqbYqdyy/XublWs2H3tV5nndnmbNe1PKB2srlqM8Pf6ShQBW10P2hgAEIAABCEBgrQggYK0VSY4DAQhAAAIQgMBLCAwol+nhZ0pyDS1YidpaFsK8neF0JZtFtUmJPO6W3TYOPHeI+rjcWO2tTUm8ulAHzvjkbMrMmj8tYHUqm2n5aojnG6vdQkdPzkiQ0rp30tm2aOVC5zc5e6tSrS7mS/k4BcVrtUpg86qBXkVvVqnuLs3z+Yc0rm4JN816vawyuNJ0VfvnF11n5+vHWr1u55iqNV/S5ublbLKytAbNczgoYVTR9dEssXFGkVYjKqF0Rr+FP68U6HJB4YvZEAcF4Ze0KmJra2NcrU/A0+V8bFUYe09nLgXbl+TaKk+vQcc4BAQgAAEIQAACl4QAAtYlwcpBIQABCEAAAhCoSDh49NkXxSsTmSxV4pFnJ+NVymy6Us1i0RHlJJ0YkoupvLCKXXd7k1aWa4gurTjYLoHjQltyOkkYqTVnVI1K0HLZ4oUKYNZznnx+Ko4qt6mkkkCLas8fbYjblWO1VUKWnWLOwvLqgy5/7OksKO9pTgKRVs6T2DUt4cZhUzPqh/xtqTRySuWQ5j+g0sLtm+cU5v5i4Hmtr5fy1i67FotoWqHQTjF1MOV2tUu4tGC4Fs0C5MISghIhdciCHHZerXJaQM3K7i/PdWpirGclYkkoVPnlnq2FJKp6xUSLhDQIQAACEIAABNY/AQSs9T9H9BACEIAABDYIAYsb/mKteCWV0a3Nl/j1jGZkTCVZy0Oc1OExCTxluWcKp7OoLvUYzNzsLZwY+yPPTMQXvjUeA4OVNBf9Wm3wxqubJRxJzFjhJ6MZCWDLm8vnLJ6cbVXB5dtavDqufjxzaFqlb9VUhuiSxkm5h775+ES89vbO6FBZYDG/UKJYUB2m3V1DoxKCNBg7jFz/1iHXWK7ReVhz2nc2OiQUdbbNp7FOK/T97fd0XdZrzmPoEkxnjDU7If10c7neWgWhe17tNBtT+aAQxK4thdi/pz2+/fR4zPq603MWEXNiV0hdaNRjhd/7WtBtrwL67ZajQQACEIAABCBQHwRe/ERRH/2llxCAAAQgAIG6JOAys+nTTh1/qbbro2WDr2iWy+WSYLc888rinR0zl6NNqizQIo/L6yxejEkk+spDk/Hkc1MxI6tTg+rcRscrEq9mo625PZpXGRnlgPUkaulcHVoNsPl07pZdVS71y0m9bJCyUtF2WhIxhbCfGCorI6yc8qzc3+HRSipJ7JATbG5+VrlNrdHf05QEv0m5iHztXLOzRUdpkPhVjWJxPok3UxK9DhydkaA1G40SufISu2Z1/9DAtI7fFps3Scm5jK2nKxeNObnFFLJuAdHi1SY/t0Zzbzeac8pK0wurGjaK9Tte0xF7thfigcfGk9NqYkoiV7OKDCV2CbdWG8wpg8s5WC0prN3XBA0CEIAABCAAgfoggIBVH/NELyEAAQhAoI4JlFT+VROvPAy7bpxZZJdKbUW8Oh7eWbt+cqgSh05UUiaRBRuXjtXGuk0h3mtVRrb85BZKpmfm46GnJuM7z0xJKJqL6yT23L6/RQ6kxnj2cCmeO1KKMWVD2aHjdCRnJzWfLKcMrFm5llbSinIAuezPzZlMFip9YAsrdlSNaOU7z7/D1i1iudSvOtsQM9rH23d1SqTS03YTjSjT6ukXZtKtXVzNEnwqc5PxyhtUDldpTiKVrxmXEzq36aarW6IsUfSJA5NaiW8u5UFNOsepIoFM+zsLShpi7NvZGCMOT19jAcvXsd1eFZ3POlCrVnAs5DXbev7wiXISBl3muLlbeV5e2XGNHXdtcqZZCGzReVMpoTjakbVne4tWHWyLU6Ozcf8DQ3F4oBJVAfb8797eHO+5pydecWNb6rN2oUEAAhCAAAQgUCcEELDqZKLoJgQgAAEI1C8Bl8udrVloqIk6Z3u9Xp976Kmp+PJDYxJonNs0H7OyvmxVmd6WXq8Cp5Xfdq/VGnRnErL7aVQr0d375cH41GeH5bZywHek8ro33tUdP/i2Pok/FrgkHkpY8fZuFlw8R3YGrXQ+LMx5jM6fsrDk8HCHwNvx5b4cG5qJk1olcFA/dmYNjpblqmqIvXIJWfSZP9IY/X355MR6+OnpGFLZ5byOZ+GpsdwYR+Se2iZuncro2iTRqku3PrYF0AkFlk/rGmpVMH6DygefOTSThDBnPc1JJHPp4HHlam2We6ujfe3dV179zyJcrflxh8oGDxyejqckFKaAdQlz7vOdN7an3Km1LJ31fDno3qWaQ6MW614UhDtUGmjRbHN3fzx1sJRWKvS2+69q1fXXUusytxCAAAQgAAEI1BEBBKw6miy6CgEIQAAC9UfAYoOFjYmpqgSLhuRQqZUOWsDYaG1Mos2XHxqXsLHgSnL5oCPRO9oa455bO5Iz6VKN2eV3xwen41P3DUvQKEs4WzBZVVUi+NmvD8erb2oLO6bsFLLoZFecbz0v3RI7dm0tysgzueLuOUdpTkKWA8NzjQuZShaYnEd1ZKAaLxyX40vuquMjcmKNzUReWtLIRDG2bSoml9KDT8xHf6dcU+pPTVTLqe+5xjk5rCJOjc2qnwok14AsGLkkckqrD85JLLO7zecflvA1onG2NDfEbMnE5fSSgDg5o8cSs3oUTr/a5vH5OrZbzELfUvGqdmyvpPjMoZJcT9OJrZ+fktvtm4+HxKSeaFNZ5Vo2v4faxL6QLyRRz320sLVFbjPPq9ns3NIsGnJqqWR3pQLlWvaVY0EAAhCAAAQgsDoCCFir48feEIAABCAAgZclMDBciYefnlJgeDmtwuZyqn17WtKXbospG60dVOlY1bVrZ7T5OCUHkgWYS9ksqDx3WKVrDvVeciILLlMSlB47MBO3qZTwerlwnj44FXmVv9mp5HK8u27oiG19uTh1asmOK7hrZ1FR5XMWoCyKDY5W4/EDU/GEsrZGde4BlVROqbxvSiKbMRw/OR/PFKeT08rurQOFXJTVHwsy0qUWSuJ068cFOYssus3IbTWvssQZlUhOyUXmkkSLWc7PevLAdBxTGHyDBNNCPpfEr0bt3NrcFDdc0xIl7du8CuOb+/j8sZnkXvNYG+Qia5ZYZnfV0jauvKtTKlc0A3OfVv+qmpfZOa36qPfCVa3FpZuv2X0LU3ZcLW9mQYMABCAAAQhAYGMQeOlf+o0xLkYBAQhAAAIQuOIEnMf0jccmw7lEXnnQX+YPypnicO3X3N6RnrvinVzjDkhrkeDSGDPLRCw7sZxZdClbo5SfVHqnPrhZ/JHxJt3OqwP5pvm4SvlI5t/fk1dZX1luqAa5dYpx+/WtykiycnTxrV3js3g2LLfVMWVqJdedxB6XD9qBVZU9qFxR6paFJ4WzTyhvvUEij8W2fK6qTDR1V5wWo6IkRrUoS+uqbUX1OZfG4RX8GjUoC1jOfzqgPK/HJV498uxUciBZNCo0zUb/JpUbqmRui0o3d8uBdLawcrsDbZQzpvPlUz1xcDpOSIR1eWOjBTW5nqZmGhV8nztjX+dR6bBJrBxPJY4LA2zKzykTbUYut0LKfrt4yuwJAQhAAAIQgEBWCSBgZXXmGTcEIAABCFxyAieHZ5N45RM1SShpTwHX82klPIdxb8S2c3M+uruaYmjYK+1ZmnFrjBuvlkB0iYdsAWlrn7O2igslbFJSfE47hnrlfHvlTV2plPEGObB6u/IpI6sgYchiT4dyq8qu11tFswuoVQKTRSELaa1aLbCtRQ4kYZjVP741gpoRLa2Md/p83se4iiov1MJ9Etaaoq83H7df2678qDaJf+7fbDq+Ip+UlVWJidJsHFFQ/iNyejnXK7m3dLyKRLIhOaF2bGmRUNqulf8UoL5MOyxX5jT+2SQ2uQvuu0xgL2l2tdk59e0nxqMgjp1yqzW4r3q+q12irI7Tfvrg5rxzSzG29hTi6EA5nnjBoprLHEN9UfbZnkoclbC3W6WaNAhAAAIQgAAEILBSAghYKyXG9hCAAAQgAIELJDArq42KrVL+Tm0XO2FkYkliQ+25jXTboZKyu29uV9lkSblMdp41xnW7CnHXTVI7LnErSNm5dmdzfN8/6os//YdTMTQyLc6NEnDy8d43bJK4kl90/2zvz4d/1rpZxPFKeHY/jUpEcvZXZ5tKGEvKyZLIFPrklbOgpZ+lzd6vWulgUWVvW/uKcecNrXH9HtX96ZrJS/BU/nxa8a9J47RY9+yhchw9NRNlH1cqkUUzO/3cChKk7ry+I27bJ/FLQlpr8UV1yg6ppeKVt7dQ5Qyupc0ZYY9oJcf/fu+JODQwo1LAiL7uXLzujk6NqSmVNfZ1a0VJnVPDTvliFtFuubY5PnX/cAqxtwvRgfkn5XZzuL9XoPRqiGbk5oUMPHZX+pkdDQIQgAAEIAABCJyLAALWucjwPAQgAAEIQGCVBDZ15pN4MCmnzMLXdGsRDcpaWjvhZFz5SjMqTcxL1GguOofptIKxyr6vZvc9W5tju5xQZa265xI9C0uXq3m1vne+rjtu2dcc33mqJNfTfNyoDKhrd7SkAP1L3Q/Pg91M6Udjt1hz7Y426Uul5JoaP30tzKh80ELm0mb5yO4sVTiqLE/lkBLfnGM1qIyrKe1n057zr0r6OS7h6qSen1cWlZv/1eapZNKGqBYpQn1afdClh63KqvLqiJOleQlGc6n00HNTLJwpGM3otaXPnJTz6r/fezKJV9MKYvdKjYcVSn/fA+Px1rs71X+tuOhVGHXrLK6yrkWHzT/6bCnKKiEtaCCzKpVsksLlbZ7VyoRPHSrGLru0JB56lUYLXG4WvryCoXPEaBCAAAQgAAEIQOBsBBrGxsaWfXw622Y8t1ICFfn3Dx8+vNLd1s32M/pkXSxm1+LP+Jn/rF7/zsRxGRXjX7vff0dPVuO5YxINVNbleq2O9lzcdHVTtKhMbDVtRq6bhw5EnByqRrNEhM1dobI4uY06Gl8iTFzoeTz//vtVKKyycxd6wnW03ULguIPRZ1WuV5Ab6OI7NyVx6uDAnBxYLvNrjJLC2Ut6bmBoVjlQWhmwpAysaZffyfkk89TS5vPq9HKNRezcrJURN8t5JbHJ5YYOJC+pjHCm3KAV9RYk0cHReeVfOQ9r4Sg2MdnNtLm3Md7+qiYJUl51UQHnXSr/a9EjHd9uq+HxuXSdtEr0tNi20CRINUt40vzrUoivPFqNT9xfSn13dpdOnUQ3C2x33SAn1q1F5WvJUXjauOUSyRNDc/H4C1UF5le10qIFs9NH1mtN+s+mt+3Lx417G+Pq7RbXTp/29I0dWJ2tV86JtbDS46xE17UTmM8c4fp+5GvfDBh/dufffwOa/EbNYKvaYqrG+LM7//4PjDlbmeuw+XP79u3b67DnK+9yNq/QlXNa8R7+5bdnz54V77dedjhw4EBd93+1HBk/81/P79/VXP/+8PrCCy9k9v3vL3D+jw9rOf87ds7H/n3VFO7d1pyP3h6vEFcTDC5utlz+9Zt/dDy+/eSkXDdzsbm7MW66pi26etqiu7dVrp+L+/PuD/DHjh2LXbt2XVzH6nQvr+43LqZFibeDg4PR3r01ra5XWIUbaMeOWYmLckiJSYvK9xzg7nMcGqjEo8qsOnSsnFYmHNLqhM6R8nb+J+/8K01frqkQI5NyYQ0W0uqDzkzr7sjrvlQki2u6htpb89HeJtdUYzmeP15WCaDFlwaJmM3JgZbXCoQWhSY1vukRlTa2NquEL6++KJeqoJUa9X63Y80lgHZS5XMSwmZGY9u2be5NPPz8sALbB5IbbEbvjZTZJSGqOtcQY1PN8dgLhdjU0y5nlsU5rY6okkN/9+3pkVB7clLuK4XXyzlmh6Az4LxK4JbN7epHISqNhdizozmVHqaTnf6nsy2XShGXPne57k9NTYX+w25s3br1cp1yXZ1nclILTuhn8+bN66pfl6sz4+PjulZnoq+v73Kdcl2dZ3R0VGXC1ejt7V1X/bpcnRkZGZFwP6/fXz2X65Tr6jxDQ0NJvOnq0n89yWA7pSWILQJ1dHTU5ejPtlBLXQ7kAjp9cZ9wL+DAWd/EF1G9X0iNq/nPzxvgAmD8/oaU3ZbV+feHN7csj9+/u9dq/BOlOTltbE9RDlJ7IZXSuaRrte1Dv/ZsPPJcaUH00MFOjlSTCNKrlfW2akU9B5lfTKv97Vqr8V9MH67EPpOlSvqbbVfUqFxDcaqaBJndKoVsb21czM1aSd+65YRz/pZL9hZaTq6qYly3t0EljW3x9AslldRNx0MSIU8qK2x80iv8uRRQopVUIJcGTitTfnBUOWpeHVH/n5ETqyBzSEEr/Y3p2mpttWCU0zFbFFDfKcdTJTpamuTIa5CTzq9pNUiJR9O6Bmf13h6brCa3VU5iWLuEoimN99TIbBJXLdYVJGA1zFWiWaKYVxm844b2+P/uG1ZA/HRyZHlhSf/kcvPqS1WB7PPxJ5+RA1AuLjsKm1WS6D7vUkj7jv6i7s+p9FHuQ+m1zvXaoxUgXSboMkuXMFYkfLUo9H5p83iuVBaWr/u1fP8vHVc93M/q+782N1kfv6//2k+NSZZuPf9uWfv7V5vjrF//WR9/7Tqoh1sErHqYJfoIAQhAAAJ1R+D44Ez89RdGY3CsLIdUPm69rkPOqFxy17RKgFhJc+C3M5B8+xf/cDIelni1tDlP6cDRssq3pAiszty19LCZuG83knXbSZX5TcvBNKWk9CblPZV0O62sqa3KK1saOr4Uivc7/Z1n6dOL97tULjpT0aqEqq2zmFTLd9qlFfnssmqTSNSgDKgjulYODcgppQwpPSNhysLXvMoZtYpi0nckSGm7KYe1zzdGi3Lda9alDm3br5yt/btaVZ7qvRviqYPTuhbKcljNqYRR2VRybVl4sgOsXasZNulYLXJneVVD515Z7JqYkgtstqq+aqXDjlJapbFN1+l3vaIz/urLlahOzqVjOGervbUhprS9jz0hFSwJYhLNejsLSTw7MdQQt1/XmlY/PDhQ0nYNcdU2ucUUPO/+FRTg7myyMR3T5YytzQsioV1mL5YzLmLkDgQgAAEIQAACEEgEELC4ECAAAQhAAAJrTODwQDl+438ci4HB6ZAmkFwdDz87qeDr7ti3py2u2r7g9LiQ0+p7vlbzqySRZUpOmv/x94Nn3c3Ol5IcL15xjnbhBFzN6f/yakeQxSyXEz53eCaekTvK950RdcM17fGu1/Qkh9GUhK6jJ8upNK5oJ5NEGa9y6NK3s4lZFq3sPlraLNTs3FxUSZ1W41N5YZNcWM7dOHqilBxVLXIyOTB9QoJWh47rgHQvUWhBK19QiZ0cTEX1q625IMdVY9x8dVts3yzxSMKQBSGX8w2Nqo9THtNcchTUnGR2hHllyOrpbKppiWLTLjGUYGehzcHqBY29ovv9WmHwmt0t8drxjvjWU5NxarQSHkmTMrNm1JcJi1oSyXyNlqSpjkvU6pRA5qD5qoLz9+9tjuv3tsaQs8AmXOoocVD1ivPaob9HpRpJqFsorexX4LwFPxoEIAABCEAAAhA4FwEErHOR4XkIQAACEIDARRL4xP2nYkCrxE1LeND3/FQa9vyReQVbl6JLOUbdHRI9lAd0Ic0liHb6WFz53DdGoiQXzdmajVd3Xt+a3DB2xjjTiHZ+Ai5Xc+mbS/TscBsYrsYLJ6bkhqqGCupiXCWFA8PDcUqOpu96ZbdK7sppNUHPi/fd2qvVFpNbKp8cR+c/44tbtMl5dMf+1tjWX4hDx6fjmUMtyQnVJYfTUbmyBlXe55UCk1tJwpID0e3iu+p0KZ5LRXdvyWv/FxcdsINp745CEqOe1qp/zQWVFKq8z1lXu1QSmQaqsfZ05pSxJaFrbDZGxqu6xiSQqaZvXKKUhdJJiVHygaXOtkiU2r2tqIwsZ18pJ0si2VS6tud0nnTEUCxWNEhAnW7y9SqBq9oQ29WvzrZGjakQh0+U9X6oyjjWGL3dTSmnrSjBze4rt7xEvitVOpg6wD9nEPD1MKB8tk9/SZlguuZu2tcm0TF/UQtEzEmw9O+kx1U2e/XmXPT3tZxxLh5AAAIQgAAELpTAhX16vtCjsR0EIAABCEAAAnHwaCU5U7wqW1Vf9v0Vf1Zf4k4oW8kh19Ny8cyoPMtuGIsf/p+Dvl1StrzZdePso2OnFP79bCmFdNsxs7zdeJWCu69tS4HcdtW0aeU5O4vqoSVG6qqdSVei2UXVKffPU89X4mtaeW9Yoo4XImqVwDIjVcvldYc0pw8+Nhl7VQq3V+HjFnH8Jd9OKQen+xgOMbcos5Jm0Wa7yhT9c6cyp5xTZTfUuISiBx+dkDtKYeu6jmZ7VV6n/Kvd25qVL6XwdAmgeZUldkkMXd66JVbdvK81iVY+nq+xbglWdpuFxCULSNW5RgV2K2j+VDkOSlyyy09VhdHX7aMtlFVaMJuU6ORFCjs0vtzWQroOhyWmzbvcUc87AN7Cn69IVR+m1t3RFEUdy0KW4uLVj8a4oc3MJHJpi9zpVQ9d0miOvkwt0LavdmnOhdPz7yoJ+P34f/3+kfjcg2OaF01aYy629OXif//RnXGzfsespMzT8/r7fzkQf/25YTkaJWDqWHbl/fqH9q6yl+wOAQhAAAJZJICAlcVZZ8wQgAAEIHDJCFjUcND23GyDygflmrGKpeaVB5slKjm0ulFCjQWDCQlZzhLy444WL2GfTyVjtc5Z9LKQMSY30ODYTHLKtKnsakbZRV75rdb6unPxKz+zd/GLpbqQyrqukB5U69Z5bz2+sYnZJF54Y2cpdSrkew1y7s977qUbmNeQgvC/+picV8q/mi7L9aQnRydV6qnbBmlEOZXwTZbm48kXqsqMmo+d2wppTi1CjsrBtEU5VD7OapoFvE0qK4zOhaNY2DlywgKUgtLlVPI5nEvlskKLV826liwgna25nHD75nz0TOd0XTkwfaE80oKohaVBOclKuo4OalXEQypdTI4qbXNqTPttWZiTIY3LwulkyWOW8Kpx98mFYxdV6LIu6/q2aFVSPpsv85z61C9HWl9Xk3KucikIvqByR1/7Hkt3h0ojdYyUnSVhzc4yN3PTIoepZNJCIO3KEvirL47G/Q+OaG4Xfnc5/Oz4idn47T89Fr/4k3tihzLcLrR9/fGJ+NR9Q5rfBXVzTsd67Lnx+A+/dyR+4QM7LvQwbAcBCEAAAhBIBBCwuBAgAAEIQAACa0jArqfbVcr3qMoFxyR4ODzbrUMh1849ch6SBYUDR2YkTlWiSXlELRIlJqYWHCv+Au8v+24u5fLqcC7dmpWAYNGiTU6tvNw4Dh23SNYht82Hf3RH2m5hrwUhyKLBem924NihUWsWV+wYutDyytp+q72dkADzxe+MR0WqYKf4V1T+5qyrSTmqLKe0iHtepXh2nnjuBiXsbJULKqfv8cbslfP8XX+tHWRX72xWLlQ1Ba0XdG4HwbvssOV02d35xt2uPDSvbOgsr3ndcSi9nVS+vpyv9axKDH0NWkCyDqZ8eF1Xc/HQc151UEKphLOjErAsMs7I1Vf29daqksVNC6tqjqrEcdirJ0aTrsVqEtZ29uejoHwuh9T7wNYthG7hnCojS3M8IUFWeVouL8xJgbPI5VJJl9wiYJ1vVi/96/c9IMHJtr9l7ajE1HGJmZrRZa+c++Gnv/iieLV0q4ee0y88GgQgAAEIQGCFBBCwVgiMzSEAUiJX5QAAOV1JREFUAQhAAALnI7Bra1HlZxIJpH5U5HaxS6a5pSkaLYRIiDg+KGHg5Ewqu3KRlZ0tDrA+peBtZw25FM3Czpi+6Fu0ckmWs7N2bJFQIEWsOtsYm+ZUpiWXyyv2d6QA71qfGqVOWKRY782iyUIZXsSIVmq8/4HxeOpIKXokyH3PW/pjh0rq7FazgHcpm8Wdk8MVZT9VUpmn3XMOZbejaEpuLIs9Lu/skuspl7OTSULiaUeRy//aNa+bVJ5nscjh5mvZLIg5d8grBZqXWaxUJLNbq137WqSzw63W3PeRUWVdSbCyyGXBsyyBy6aow8dnJJrOx2Zx8MqMExIVPe5iitqqxlSloOtV5Yu6Znsnm2JcJY9TpYrKD4vJQZayktTXQ1rMwOWDvh5dBuu+uw/OD/O1fXK4IVzu2Cr34SGd00H6O7Y0S+jN63wq3ZSI2CzB1wH5Fyra1cbH7cUTaJDDzleKLoUzm36P+b2wktag6+qsTY5GGgQgAAEIQGClBBCwVkqM7SEAAQhAAALnIXD0eCVc6relUeKDxCh/FXSm0rMHp+MfvVL1YVIJbDzyamwyYKUv8/6a57Kvk8PlVGblL/vT+kJvEcVf4Hu7cnHD3rbkUHEpm7WIHXJ03XZdW7otp/NY5FCJoo5ZD80C1vNHZ+Kjf3ZMId/TEucWnDoPKPvpg9+7Ne66sT25eixkXapmJ5JZ9yhT6rgWeLRg5YyobokzRyQ0NqhPne0LwlFBYkpeE2mBct+eFrnqNC8q32yVg+5sWVRr0WfP5WrH7zEtLzW0cNQk15OdWR60xSKLVH5Yna9KxFLZn1xYJa9OqOt0XoJDo2opneFmcau7vaBxyxnYtVCu2trs1TVbUjlsk14vyW3V3GbxVkH4YlxzIs5ozsfkYHtWKz0e02qOjdp2ekblibp+/R6wiHZUqyi26Nj75EC757bOuHpnMXZtkSh8gc6zteCe5WO87w3d8cizExIUJW6eBuG5uXZXUcH8F+6+8q7veV1PynKrlRD6uQb5/W68iiB3s6BBAAIQgMDKCCBgrYwXW0MAAhCAAATOS6AidaooIcklUU36Am9hYEwOn4eeHI+P3dsQu7VK3Ly+yDsM+5lD0wp6n5VQk4+dcp9YsGiTMOLmL42TEgIGBityZ1WSK2tTR1EhyE2plMwZSP5yb5OD79dbK4nP331tJI6cdMnaaceHbu2G+ov7BuM6iUQWQFYr4JyLi8sB/ePj37KvQ5zLElWkpqgPrXINvVPP9WkFtgcfndJcWlTxF++OuPOm1mhdwtv7e77Xc3O4/JSupZogYYHo1be0x/GTVfGfSqV+vt4cbyWdLiqq/TsiFs0qBxSilHtVkODVnMoYc7G5tym29RWTM805YN1yCLr8z0KUyy8LKqv0OdwkhaUsOJdZTskJdlzB8c51c87YlN4DAxIK/R6YVunmKV3rFnd9TR+VqPmMhK5/+o7eJBK2ejlF2iUn8GattvnEgen41BeGJGBW9R7MxXW7i/EffnZvcpCupAN33dgR/8s/2hKf/OxActX5WPv2tMb/8eM7V3IYtoUABCAAAQgkAghYXAgQgAAEIACBNSZw9fZifO1h5/soq0qiweCwgsGVYdWUq8anvzqcyuR27yjGkJ6fcKaMvqx3yOnS2DgvsSTiDXd2StBxHpRyh1TiduREJWUKVaQAuJzQ1TcO9N6zvXnFJWVrPNSLPpzdV3Y+DUiksOtnsYmFRRYHiI/ox2VqdmZdCleZy+j8YxFr99Z8vP2e9vjOk7NRbOmQeNas8sy29Pqb7+yWO8uuulxs0qp/nleXeFqQsbPJ47hSzdeHw7abUpbUufthIdArLTpA3TlULk296eq26Ja77/f+QosJlEoLYfUSr5okMjTqIvP4LD65oCwvgc6utLyEqQ5dg7v6i3HXTR2pbNIOr9qKgs4Pc5lZIb9Q9mgunjuXzpYlXvm69rbzysZqkSA2Oqb+iL/LJId935OvZkGzqrfG8ZOluP8b43HNbhw7C2Qu/b+ez5/+ga3xwe/bEk/INbpLmWddXal+9KJO/mPv641/9p6eeOCJSmzr1ntNv7doEIAABCAAgYshgIB1MdTYBwIQgAAEIPAyBO68qT2+8th4jEsMmSyVQ+ad5GLxivQWPgbKFa0qqBBrlRnm9G3RX+pnGqtypuj1uYa4SuJWW2uTViicjaGJOe3r8q0G5WpZQJBAICeQxZQrKZy8zPAv6CV/SbYrqEulaLnGUuKzdMeiPqG06jvzgsi09JW1vW+327gynnyeLb3FeMV1ubjmqk1yE73oqLK7ba9EyaXN7C/3aolLz+/7owpXt8vv0IlyfOOJSQXON8S7X9sR2zafXSBweekm/Viw83gtJLUUW+M9r++NT33uZFpNUOlrWk0wp1K+OYlWKq2UWHVMjqkJbXtU1+41cg9eK9fffpWAOXjdixa4WVi1ENXRplUGtbKmXVlJqDrt+eqR8GeOdhM2SNHKu3yx0pgWMcg1qsRQgtXpherS8dI/OrSfPzk8I0fXwnlefJF7l5qAr/Gb16jUr1FzfveNZ76HLnX/OT4EIAABCGw8AghYG29OGREEIAABCFxhAhYHfuSdW+MPZ0/ECeX5TDTqS7u+fyenkZ0lcp/YYTI+NR/N0hpcllXRc15ZsDBejqcOlWJzdz6mJCKMy7UyOFRJIdzNcrbsVAbNtXKj2J3i5uOc1hAWnqiTfy3EFSWQvPlV3fHYCyUJJGXJLwvN7qBX3NQdfT0S8lpeFJIuxdCaJaTkGpuSEGSvUYdWxlsqXl2Kc67FMS1c+ee+B4fjk/cP61qqJqHz3i+ciu/5rt543xstwp1d5PT16eaAd5cF3nNLp0r1puPg8YmYmpRzShdUqxTEopYSHBmfSWWBswpVt/B18EQp7p6zQLbgLly6aqBdXjkdz5rVhLboUnaYHWItKkP0dhZeOzSfU3IbHjo+nxx2Pcobc5B+SSqv3Wx2fbnZteVe+tpOKx+uMHspHYR/IAABCEAAAhDYUAQQsDbUdDIYCEAAAhBYLwR6Ohvjjv0t8dgzE8q6qsTcacHJGUMOBveXdWViSxRYKLWak3gypQDrro7GmNEXfJesjckZ9MRzpZR/5a/zdkS4nFALvqV8qFMjlSRgWQw6l1ixXnicrR9dChK/+5Y2hUX3xZ/fNyy3WSUJJ296RVf80DuVeyTBbq1X9jtbP8zPpXFFhT9NT55ti/X3nEswBzX/f/OlUTmddB2oi14UYKZcjU99fjhuu7Yldm5tiW6F0/taO1dzdtfW/sb4/rf0xF9/oRrPHZQbrdCc3IFeROCISjytSOXEqEFlhdMzs/Hl70zFK67vSq7A9taXHtlOK//YkWVRa2nbqtysdrkLH1JF4NCYVjiUQLt781wcG2qJbz02plu9Uew4VJ/d705lw/3QOzalnLKlx+E+BCAAAQhAAALZI4CAlb05Z8QQgAAEIHCZCOzf2xL33NqZwrBHVYJ12lwiMUDlcxJMWvTlf05CRHLESIHwqnZ9XXm93iiX1mx88/FJlW9J/NJrXsWwWY6lwZFyHD7WEIcGWhWsvKAeWMwYnahqpcL8ZRrZ2pzGIoVLz977pr549xt6JZPYDbU2x97oR/E1c2q0mlaw9FgtXtVaWSWqjzw3Hdu1KIBdWhfiYturFQT/+fv64+BhBbW39ceTcsXd/+CYdVOV+S2U8jn7y5YoZ7N5ZcHzhesvF6/cv+SokvPrtbd3yXWl/C45u3ySimyJx05tir/8wnB864kpXf9Vra7ZHP/kLb1x963ttaFxCwEIQAACEIBAhgkgYGV48hk6BCAAAQhcWgJ9KgO8+9aOGJaz6L5vjMWEnFh2YPVoxba+TYVQpFVyqcxLuuntdLmcQ8XtOmmML39rPAYlUNRKqmbluqrmLPG43EolWlPJqJIEAY/CQecu1ypI5KrHVitrW03fqxLyxpW/NKmV05ynZPdWu0RBZ1W1qnytlte0mnNcrn09l86RcmtWVthyJ5pXB7Q7SSFqujC0kZUhOZcaJGz5fy7PdHO+2oU283H55FYJXyPK19rU1ZTEVR8v6YoWUnXHiwHmdccZZqtpLi0MuewWmvK5JMDedE2LVr5bKI8sysXleaNBAAIQgAAEIAABE0DA4jqAAAQgAAEIXCICLvnbt7s5+ns2xytvatMKd6UYVVlgpxwoDgxvlBiV1zaWKbwK28j4bPrC7hDsUWUaebnBWgWW9Yk5iTIWqhyuXZR24XB3h8DXmoUwu1my2MzlxEhVKz7OxB//7VAcPjGTtJ27xP1tr9kU/RITuxUkXg9tSgLOyeGysqc0o5pjlzdu7c2f4aSyu+mGq4px9Y7mePLAZLqWnF3VKIW0X9fWzfsW3Hm+Vlba7PbbtaUQ99zWEc8drsTA0FRyAVrIcqnqDVe1Rf+mpnR/pce+kO2dleUfGgQgAAEIQAACEFhKAAFrKQ3uQwACEIAABNaYgPOV+nvy+umO19/RHeNytoyp/CqtPCiBwkHtysqW68T5RXZQRXzpoXE95y/xOYlWsyn03bqUtaq8HCs3X92SAriXlmh5lcKCa70y2qbLczFwaiZ+6+MDyg8rJQqWbu57wCvcNcb3KePJpXSej/XcHJQ+NKqcM10XtbZQXqcVAHc2Lzru/JodUL/yUzviV//7gEQsld3JgbZFwtIPvL0vlZO6xM8ZVxfTXNp5w97WeP97G+MzXxuVoFZRPltD3HJte7z39c6kWt8cL2bM7AMBCEAAAhCAwPomgIC1vueH3kEAAhCAwAYj0NGeC//UWkXOq2mVB7o0y24r5wvdfO1sPPV8SY8r4UXdvDqho4KaC03xpju64jW3d8fmTS8ew+V3HSr9WmLGqh3+vLcW0rwqXEUnqCqHyKVr9dgmFQb+wKMTcWpM9ZWnm0dSkQD48NOT8YZXdKTSzXUvYKnTdl5ZzJyRKGdTXZNcVC1FBajr8fLcqZbmfPzbn9ipkkmXTkrU0rZ24tn955+Lbb6mXEJ4983t6UfZ8NJQ67dE9WI5sB8EIAABCEAAAuuHAALW+pkLegIBCEAAAhkkkJdDJr8ke92rt/V1dyiUfTa+9p2xGJcTxmHXdmm95dVd8Za7e6NLGVkWq5yPZQFqNUKFz+MQeDeLaROlBaHkhWPl+PaT4xLKivFqleG1KEtqPTdhUAmmw6CcE5Z0n8XuujzTGefLc6QWN1hHdyxiOnh9SqJlrc1KzIr5XBpX7bnltzkpoJ1ttWcvXriqHaF2WxNFXbKa1fLUGgtuIQABCEAAAhC4sgQQsK4sf84OAQhAAAIQeAmBRiW5v09lWldvK8TzxyoSXiJuva49dvYvUbq0l11bqxEVqrN2Xi2IV7VOOEvqI398LL75xGQqX7Rq8snPFuN/+/Gdyu1Sevc6bS6n27klH/nHGtXvubQqn2Ucl1l2tDbFni3FVQl9l2vYdj5ZmLSA5VB1O8YcmG4XlhcAoEEAAhCAAAQgAIGsEuCjUFZnnnFDAAIQgMC6JuDVCC1ave+NPfGu1/a8RLxai85bKFneHn5uJr7+6PiCeOUXpW8dHSzH7/3FCQXInyl2Ld/3Sj4uSOh5053dKr/sCLvYFpxtjdHWVogffmd/bOquj/9mN64S0kJ+XmH+uSRkOdB/Xkpch/K7nHNGgwAEIAABCEAAAlklUB+f5rI6O4wbAhCAAAQgcAkJaHG75PJxGWKtPX1Y923Dci1bren1Q8emUzZTa/OS52uvr4Nbr1pXmW2KD35Pf3z14ZZ47mg52vXcu9/QFTs3N6+DHp6/C3bEOefKAf6d7RFHBsrxwvEZZZ81JGGuR+Wk3Z18dDs/SbaAAAQgAAEIQGAjEuBT0EacVcYEAQhAAAIQuAACLq9rlcjjAPRaK1jUsv3L1qslrSmfkzNofYpXtW52tuXSeN71+h6VCzaeLrGsvbr+b2dPO9zalDd275dG4jNfGY6RiWrMa3ramofj/e/pj3/ytr4oqFySBgEIQAACEIAABLJGgE9AWZtxxgsBCEAAAhBYQsAClp09Fk3aW3PxxruK0d5y5n/fcuXaPTe31kWGlAPtiwrGX8gHWzLQOrhrQdHt1HA1Pvv1sRgercaMFlX0CoDDCtv/fz9xIu775lgdjIQuQgACEIAABCAAgbUngIC19kw5IgQgAAEIQKCuCHh1PgtZLcqOumpLLj74fZtja39zKmVrby3E21/dG//03Vvqakz12FmLb56DJw5Mxuh4OQlXS8dRUXnhJz47GDO6pUEAAhCAAAQgAIGsETjzP7FmbfSMFwIQgAAEIACBlxC455bO8M+sUt5zl8DKdGqoEl95ZEILHM7Ha27pik09qlukJQJ2wbW0NYqM3VhnlnF6g5mZSgyOVWN73/pdETINhH8gAAEIQAACEIDAGhNAwFpjoBwOAhCAAAQgsFEIXArx6r5vjMWf/O3JmJ6pKkA+UtbTP3t3X9x9a+dGwbbqcbz25u74ndYTMV3SCoS1o4lVQWHuWza1RItuaRCAAAQgAAEIQCBrBCghzNqMM14IQAACEIDAFSIwPlmNP/27oZicqsjdNZ8WO5wqVeKP9dyShRCvUO/Wz2l7unLxoR/eFW0KpbcBrlEGtaIMV3u3t8Qrb+qMns78+uksPYEABCAAAQhAAAKXiQAOrMsEmtNAAAIQgAAEsk7g0QPTMTY+vegqmpOIZYFmbHImnjs4E9fsKWYd0eL43/TK9rh9/3Xxf3/8RBw/OR1bN+Xj1bd2xFvu7lrchjsQgAAEIAABCEAgSwQQsLI024wVAhCAAAQgcIUIzKkWLq/Kt8aGXMzOzy72ws+H8p5aO/lIsgjl9J1uOa3+7U/sWP40jyEAAQhAAAIQgEAmCVBCmMlpZ9AQgAAEIACBy0vAJYNb+guxY8uZ4eMuHbx2d1tsI8j98k4IZ4MABCAAAQhAAAJ1RgABq84mjO5CAAIQgAAE6pFArrEhmguN8YHv2Rw37GmL5qJW22vOxy3XtsXP/eDmehwSfYYABCAAAQhAAAIQuIwE8OtfRticCgIQgAAEIJBVAo36T2YtxQaFtTfFB79/a1Qqs9HU1Bg9HU0KKOe/p2X1umDcEIAABCAAAQhA4EIJIGBdKCm2gwAEIAABCEBgVQTaWnKRl2hVrsxJzcpFs0KxmpoUjEWDAAQgAAEIQAACEIDAeQggYJ0HEC9DAAIQgAAEILB2BAoSrQr53NodkCNBAAIQgAAEIAABCGSCAJ79TEwzg4QABCAAAQhAAAIQgAAEIAABCEAAAvVLAAGrfueOnkMAAhCAAAQgAAEIQAACEIAABCAAgUwQQMDKxDQzSAhAAAIQgAAEIAABCEAAAhCAAAQgUL8EELDqd+7oOQQgAAEIQAACEIAABCAAAQhAAAIQyAQBQtwzMc0MEgIQgAAEILC+CEzPzEW5Oh+NWoSwudgYTTlWI1xfM0RvIAABCEAAAhCAwPoigIC1vuaD3kAAAhCAAAQ2PIGxydmYKc8tjnNa97s7mhCxFolwBwIQgAAEIAABCEBgOQEErOVEeAwBCEAAAhCAwCUjUJHrqiZe+XayNJdcWOVKxPb+/CU7LweGAAQgAAEIQAACEKhvAghY9T1/9B4CEIAABCBQVwRmZ+fDItZnHxyNez8/GAND1cjnmuKVNxXjn79ve+zeVoh8E+WEdTWpdBYCEIAABCAAAQhcBgIIWJcBMqeAAAQgAAEIQGCBQE5ZV199eCL+6N7BODVSPo2lHJ/5SjkmZ47H//r+7bG9DycW1wsEIAABCEAAAhCAwJkEWIXwTB48ggAEIAABCEDgEhJoVGr7Nx8fi+HRmni1cDInYn3j4bEYOFFJDq1L2AUODQEIQAACEIAABCBQhwQQsOpw0ugyBCAAAQhAoH4JzMfJkfmYm3/pCMp68qnDk0EB4UvZ8AwEIAABCEAAAhDIOgEErKxfAYwfAhCAAAQgcBkJ5OTA2rcrn4Lbl5+2qOyrPdvbookMrOVoeAwBCEAAAhCAAAQyTwABK/OXAAAgAAEIQAACl5fAj79va/T1NEXDaauVb5pyETfva5O4Vbi8neFsEIAABCAAAQhAAAJ1QQABqy6miU5CAAIQgAAENg6Bnu6m+MP/c3/cc3Nb9HfnY0tfId79uu74pZ/cFZ1tUrJoEIAABCAAAQhAAAIQWEagYWxs7CwpFMu24uGKCVSr1RgaGlrxfutlh5GRkeju7l4v3bns/WD8zH9Wr//5+fkYHR3N7Pt/bm4uxsfHo6ur67L/3lkPJ/T4JyYmorOz87J1pzQzF/NKvbILq5BviNwV/E9rs7OzMTU1FR0dHZdt/OvpRP7sUiqVMjv+SqUSMzMz0d7evp6m5bL1pVwuhxm0tbVdtnOupxN57v07oLW1dT1167L1xeP334CWlpbLds71dKLp6enUnebm5vXUrcvWF//ub5AtOqvj99/+XC4XxWLxsjFfyxM1NTXFpk2b1vKQ6/ZYTeu2Z3XescbGxrr+A+AvsFn9A+ZLj/Ez/1m9/i1gWcDJ6vj94X1ycjKz468JOJdz/tfTdyV/efeXmMs5/vX0cacmYGR1/P7s5vdAVsfvL6/+G5DV8Xvs/h2Q1fH771+WBSy/992yOv/+Dxj1/v11NX9P/d63CFSv82/xLSsNAesSzbR/AdTzf8E9ceJEXfd/tdPK+Jn/en7/rub69wf4U6dOZfb97w+ww8PDmR2/P8DJmZ3Z8VvAybIDy+Kdf7L6+89fAPweyOr4LWBZwMjq+P33r94/v6/m77/n3iJGVuffY/c1kNXx+++ffwdmdfz+22f3VVbHv5rfHZd73yto1L/cQ+V8EIAABCAAAQhAAAIQgAAEIAABCEAAAvVIAAGrHmeNPkMAAhCAAAQgAAEIQAACEIAABCAAgQwRQMDK0GQzVAhAAAIQgAAEIAABCEAAAhCAAAQgUI8EELDqcdboMwQgAAEIQAACEIAABCAAAQhAAAIQyBABBKwMTTZDhQAEIAABCEAAAhCAAAQgAAEIQAAC9UgAAaseZ40+QwACEIAABCAAAQhAAAIQgAAEIACBDBFAwMrQZDNUCEAAAhCAAAQgAAEIQAACEIAABCBQjwQQsOpx1ugzBCAAAQhAAAIQgAAEIAABCEAAAhDIEAEErAxNNkOFAAQgAAEIQAACEIAABCAAAQhAAAL1SAABqx5njT5DAAIQgAAEIAABCEAAAhCAAAQgAIEMEUDAytBkM1QIQAACEIAABCAAAQhAAAIQgAAEIFCPBBCw6nHW6DMEIAABCEAAAhCAAAQgAAEIQAACEMgQAQSsDE02Q4UABCAAAQhAAAIQgAAEIAABCEAAAvVIAAGrHmeNPkMAAhCAAAQgAAEIQAACEIAABCAAgQwRQMDK0GQzVAhAAAIQgAAEIAABCEAAAhCAAAQgUI8EELDqcdboMwQgAAEIQAACEIAABCAAAQhAAAIQyBABBKwMTTZDhQAEIAABCEAAAhCAAAQgAAEIQAAC9UgAAaseZ40+QwACEIAABCAAAQhAAAIQgAAEIACBDBFAwMrQZDNUCEAAAhCAAAQgAAEIQAACEIAABCBQjwQQsOpx1ugzBCAAAQhAAAIQgAAEIAABCEAAAhDIEAEErAxNNkOFAAQgAAEIQAACEIAABCAAAQhAAAL1SAABqx5njT5DAAIQgAAEIAABCEAAAhCAAAQgAIEMEUDAytBkM1QIQAACEIAABCAAAQhAAAIQgAAEIFCPBBCw6nHW6DMEIAABCEAAAhCAAAQgAAEIQAACEMgQAQSsDE02Q4UABCAAAQhAAAIQgAAEIAABCEAAAvVIAAGrHmeNPkMAAhCAAAQgAAEIQAACEIAABCAAgQwRQMDK0GQzVAhAAAIQgAAEIAABCEAAAhCAAAQgUI8EELDqcdboMwQgAAEIQAACEIAABCAAAQhAAAIQyBABBKwMTTZDhQAEIAABCEAAAhCAAAQgAAEIQAAC9UgAAaseZ40+QwACEIAABCAAAQhAAAIQgAAEIACBDBFAwMrQZDNUCEAAAhCAAAQgAAEIQAACEIAABCBQjwQQsOpx1ugzBCAAAQhAAAIQgAAEIAABCEAAAhDIEAEErAxNNkOFAAQgAAEIQAACEIAABCAAAQhAAAL1SAABqx5njT5DAAIQgAAEIAABCEAAAhCAAAQgAIEMEUDAytBkM1QIQAACEIAABCAAAQhAAAIQgAAEIFCPBBCw6nHW6DMEIAABCEAAAhCAAAQgAAEIQAACEMgQAQSsDE02Q4UABCAAAQhAAAIQgAAEIAABCEAAAvVIAAGrHmeNPkMAAhCAAAQgAAEIQAACEIAABCAAgQwRQMDK0GQzVAhAAAIQgAAEIAABCEAAAhCAAAQgUI8EELDqcdboMwQgAAEIQAACEIAABCAAAQhAAAIQyBABBKwMTTZDhQAEIAABCEAAAhCAAAQgAAEIQAAC9UgAAaseZ40+QwACEIAABCAAAQhAAAIQgAAEIACBDBFAwMrQZDNUCEAAAhCAAAQgAAEIQAACEIAABCBQjwQQsOpx1ugzBCAAAQhAAAIQgAAEIAABCEAAAhDIEAEErAxNNkOFAAQgAAEIQAACEIAABCAAAQhAAAL1SAABqx5njT5DAAIQgAAEIAABCEAAAhCAAAQgAIEMEUDAytBkM1QIQAACEIAABCAAAQhAAAIQgAAEIFCPBBCw6nHW6DMEIAABCEAAAhCAAAQgAAEIQAACEMgQAQSsDE02Q4UABCAAAQhAAAIQgAAEIAABCEAAAvVIoKkeO10PfZ6bm4tSqVQPXT1rH+fn52NycvKsr2XhScaf7fn3+zer17+v/SyP32OfnZ3N7PxXq9VMj79SqYQZZPX9Xy6XMz3+6enpzI/f74GsXv8zMzPh90CWx5/lv3+ef7eszr+v/cbGxsyO37/76nn+PXctLS1pDBv9HwSsSzTD/hI0Pj5+iY5+6Q/rP2D13P/VEmL82Z7/en//rub6rwlYWX3/e/xZfv/72reAk9X599xnefwee9bH7y9xWb3+/QXOP1kdv+c+y7//LeD4b2BW5782/tV8hqrnfT3+hoaGdA3U8zgutu8ev//++XdAPbZ8Po+AVY8Tt5763NTUFFu3bl1PXVpRX/xfH+q5/ysa7Fk2ZvzMf1avf394tXs0q+P3Bxd/iMnq+P3l1R/gsjp+f4E9ceJEZsdvB9Lg4GBmxz81NRWjo6OZHf/ExERyX2zZsuUsn4w2/lNjY2Pp939/f//GH+xZRuhr37//e3t7z/Lqxn9qeHg4iTebNm3a+IM9ywj9uz+Xy0V3d/dZXt34T508eTKKxWJ0dnZu/MHW+QjJwKrzCaT7EIAABCAAAQhAAAIQgAAEIAABCEBgoxNAwNroM8z4IAABCEAAAhCAAAQgAAEIQAACEIBAnRNAwKrzCaT7EIAABCAAAQhAAAIQgAAEIAABCEBgoxNAwNroM8z4IAABCEAAAhCAAAQgAAEIQAACEIBAnRNAwKrzCaT7EIAABCAAAQhAAAIQgAAEIAABCEBgoxNAwNroM8z4IAABCEAAAhCAAAQgAAEIQAACEIBAnRNAwKrzCaT7EIAABCAAAQhAAAIQgAAEIAABCEBgoxNAwNroM8z4IAABCEAAAhCAAAQgAAEIQAACEIBAnRNAwKrzCaT7EIAABCAAAQhAAAIQgAAEIAABCEBgoxNAwNroM8z4IAABCEAAAhCAAAQgAAEIQAACEIBAnRNAwKrzCaT7EIAABCAAAQhAAAIQgAAEIAABCEBgoxNAwNroM8z4IAABCEAAAhCAAAQgAAEIQAACEIBAnRNAwKrzCaT7EIAABCAAAQhAAAIQgAAEIAABCEBgoxNAwNroM8z4IAABCEAAAhCAAAQgAAEIQAACEIBAnRNAwKrzCaT7EIAABCAAAQhAAAIQgAAEIAABCEBgoxNAwNroM8z4IAABCEAAAhCAAAQgAAEIQAACEIBAnRNAwKrzCaT7EIAABCAAAQhAAAIQgAAEIAABCEBgoxNAwNroM8z4IAABCEAAAhCAAAQgAAEIQAACEIBAnRNAwKrzCaT7EIAABCAAAQhAAAIQgAAEIAABCEBgoxNAwNroM8z4IAABCEAAAhCAAAQgAAEIQAACEIBAnRNAwKrzCaT7EIAABCAAAQhAAAIQgAAEIAABCEBgoxNo2ugDZHwXR6C3t/fidtwgezF+5n+DXMoXNYwsX/+NjY3R09NzUdw2wk65XC66u7s3wlAuagwef1dX10XtuxF2ampqyvT48/l8dHZ2boSpvKgxFIvFaGhouKh9N8JOzc3N4fdAVpvHPzc3l9XhR0tLS2bH7oG3traGPwNltbW1tYU/A9DWP4GGsbGx+fXfTXoIAQhAAAIQgAAEIAABCEAAAhCAAAQgkFUC2ZVZszrjjBsCEIAABCAAAQhAAAIQgAAEIAABCNQZAQSsOpswugsBCEAAAhCAAAQgAAEIQAACEIAABLJGAAErazPOeCEAAQhAAAIQgAAEIAABCEAAAhCAQJ0RQMCqswmjuxCAAAQgAAEIQAACEIAABCAAAQhAIGsEELCyNuOMFwIQgAAEIAABCEAAAhCAAAQgAAEI1BkBBKw6mzC6CwEIQAACEIAABCAAAQhAAAIQgAAEskYAAStrM854IQABCEAAAhCAAAQgAAEIQAACEIBAnRFAwKqzCaO7EIAABCAAAQhAAAIQgAAEIAABCEAgawQQsLI244wXAhCAAAQgAAEIQAACEIAABCAAAQjUGQEErDqbMLoLAQhAAAIQgAAEIAABCEAAAhCAAASyRgABK2szznghAAEIQAACEIAABCAAAQhAAAIQgECdEUDAqrMJo7sQgAAEIAABCEAAAhCAAAQgAAEIQCBrBBCwsjbjjBcCEIAABCAAAQhAAAIQgAAEIAABCNQZAQSsOpswugsBCEAAAhCAAAQgAAEIQAACEIAABLJGAAErazPOeCEAAQhAAAIQgAAEIAABCEAAAhCAQJ0RQMCqswmjuxCAAAQgAAEIQAACEIAABCAAAQhAIGsEELCyNuOMFwIQgAAEIAABCEAAAhCAAAQgAAEI1BkBBKw6mzC6CwEIQAACEIAABCAAAQhAAAIQgAAEskYAAStrM854IQABCEAAAhCAAAQgAAEIQAACEIBAnRFAwKqzCaO7EIAABCAAAQhAAAIQgAAEIAABCEAgawQQsLI244wXAhCAAAQgAAEIQAACEIAABCAAAQjUGQEErDqbMLoLAQhAAAIQgAAEIAABCEAAAhCAAASyRgABK2szznghAAEIQAACEIAABCAAAQhAAAIQgECdEUDAqrMJo7sQgAAEIAABCEAAAhCAAAQgAAEIQCBrBBCwsjbjjBcCEIAABCAAAQhAAAIQgAAEIAABCNQZAQSsOpswugsBCEAAAhCAAAQgAAEIQAACEIAABLJGAAErazPOeCEAAQhAAAIQgAAEIAABCEAAAhCAQJ0RQMCqswmjuxCAAAQgAAEIQAACEIAABCAAAQhAIGsEELCyNuOMFwIQgAAEIAABCEAAAhCAAAQgAAEI1BkBBKw6mzC6CwEIQAACEIAABCAAAQhAAAIQgAAEskYAAStrM854IQABCEAAAhCAAAQgAAEIQAACEIBAnRFAwKqzCaO7EIAABCAAAQhAAAIQgAAEIAABCEAgawQQsLI244wXAhCAAAQgAAEIQAACEIAABCAAAQjUGQEErDqbMLoLAQhAAAIQgAAEIAABCEAAAhCAAASyRgABK2szznghAAEIQAACEIAABCAAAQhAAAIQgECdEUDAqrMJo7sQgAAEIAABCEAAAhCAAAQgAAEIQCBrBBCwsjbjjBcCEIAABCAAAQhAAAIQgAAEIAABCNQZAQSsOpswugsBCEAAAhCAAAQgAAEIQAACEIAABLJGAAErazPOeCEAAQhAAAIQgAAEIAABCEAAAhCAQJ0RQMCqswmjuxCAAAQgAAEIQAACEIAABCAAAQhAIGsEELCyNuOMFwIQgAAEIAABCEAAAhCAAAQgAAEI1BkBBKw6mzC6CwEIQAACEIAABCAAAQhAAAIQgAAEskYAAStrM854IQABCEAAAhCAAAQgAAEIQAACEIBAnRFAwKqzCaO7EIAABCAAAQhAAAIQgAAEIAABCEAgawQQsLI244wXAhCAAAQgAAEIQAACEIAABCAAAQjUGQEErDqbMLoLAQhAAAIQgAAEIAABCEAAAhC4nAROnDgR5XL5cp6Sc0HgJQQQsF6ChCcgAAEIQAACEIAABCAAAQhAAAIQqBH48R//8XjkkUdqD9f09i//8i/j2WefXdNjXoqDjY+Px+/8zu/Ez//8z8dv/uZvxsGDBy/FaTjmyxBAwHoZOLwEAQhAAAIQgAAEIAABCEAAAhCAwKUhUK1W42d+5mfi937v9y7qBE899VT87u/+7kXtu5KdDh06FK9+9avj7//+72P//v3x7W9/O97whjfE5z//+ZUchm1XSQABa5UA2R0CEIAABCAAAQhAAAIQgAAEIACBlROwAGQh6NOf/vTKd9Yex48fj3/4h3+4qH1XstMv/dIvxXd/93fHn/7pn4bdaH/wB38Qv/3bvx0//dM/HXNzcys5FNuuggAC1irgsSsEIAABCEAAAhCAAAQgAAEIQCArBF544YUk4nz1q19dHLJdVF/5ylcWH9fuPP3000lgqj0+263LB3/oh34odu/eHd/5znfO2GRsbCyeeOKJmJ+fj29+85vxZ3/2Z+Esrlrzaw8//HAMDg7Gl770pRgeHq69lPK6LI59/OMfj2eeeWbxed8533HP2Pj0g4ceeii+93u/94yX3vrWt4bLCg8cOHDG89/61reSqGYutfb888/HkSNHag/TrYWvpRz9pPl6nJ/97GejVCotbn+h+y/usEHvIGBt0IllWBCAAAQgAAEIQAACEIAABCAAgbUiYLHpAx/4QBKTPvShD8W//Jf/Mh26qakpPW/Bamn7yZ/8yZfNibIwZffUm9/85nj3u98dPv7S9uijj8Yv/uIvpmP/+q//enzsYx+Le+65Z1EU+8IXvhD33XdfEob++I//OAYGBtLuFrTe8Y53xH/5L/8l5Xb9wA/8QHzkIx9ZPPT5jru44ZI7Fth8rqUtn8/HN77xjdi1a1d62hlhdpP9xm/8Rvz5n/953H333YvCnksOf/mXf3np7vH1r389/t2/+3eLz/23//bfksvL23qs5nL48OH0+oXsv3igDXynQerj/AYeH0ODAAQgAAEIQAACEIAABCAAAQhAYBUE3vve90Zvb28qnfNhpqen47WvfW16fMstt8SHP/zh2LFjR/yrf/Wv0lksIjkz6sknn4zGxrP7Zr72ta/Ff/7P/zm5pCzU2OHk52rNri4LWxaivv/7vz897bysq666Kv7Nv/k36bFdVi7l+5M/+ZPabvELv/AL6Zw1cWh0dDRe+cpXxr333hv79u1LotL5jrt4sNN3HnjggdSHn/3Zn42f+qmfira2tuWbhBlZLPvhH/7h9Npf/dVfpXN+9KMfjYmJibjrrrviscceW+ThssRt27bFBz/4wSS+vepVrwqfZ/PmzWl/s3n88cdTxtf59n9JZzboE2e/kjboYBkWBCAAAQhAAAIQgAAEIAABCEAAAisnUHNcec/m5ubkFrILys3izd/+7d+m+/7HYedve9vbFsWaxReW3LHj6l3veld6ZufOnVEoFGK5i+uGG25YFK+8oUWz5SWBSw6Z7jqnynlVtdbV1RWvf/3r4xOf+ETtqVjpcS2AeX+7sG699db4tV/7tRgaGlo8nu/YlVYT2vzYDiqXArq1t7eHhT4LVLX2d3/3d0mg82OzuOOOOxbFKz/3vve9Lzm5XGp4vv29fRYaAlYWZpkxQgACEIAABCAAAQhAAAIQgAAEVkFguZPKjqujR4+mI7q0z+LTyMhIemxxpiZOneuUdkS9853vXHzZ29u1tLRt3bp16cPw48nJyTOeW/rAr83MzCS309Ln3/jGN6Z8qdpzKz2u97PA5P79zd/8TZw6dSpe8YpXpOyt2jEtWBWLxVTSaEeYhbRKpVJ7Od7znvekff2E87BcemmGbs6+etOb3pTu1/659tpro7+/f5Hxy+1f22ej3yJgbfQZZnwQgAAEIAABCEAAAhCAAAQgAIE1JtDS0pJKCX3YXC6XHFd2Xtkx5FB1Czrnas6LclmcxaA//MM/TD8Wnz71qU+da5cLet7HdKmj+7O0uVTPr61Fu+666+I//sf/GM6s+sEf/MEUGO/jOpDdgty//tf/OglcFtKWNgt0n/nMZ9JTS91XfsLB8lu2bFm6ebq/ffv2RcHu5fZ/yY4b9AkErA06sQwLAhCAAAQgAAEIQAACEIAABCBwqQhYdOnu7l48fM0h5GBzl9m1trYuvrb8joWq2267Lby6Xu3HoejOwqq5upbvcyGP3R/vv9ylZXeYha2Lac7z+tVf/dWX7OrAdudVeYVCN2dZ/dzP/VzK43K55Y/92I+dsY/dVM7OOnToUHz6059OjqzaBps2bUp5YbXHvnXI/bPPPrvY75fbf+l+G/l+00YeHGODAAQgAAEIQAACEIAABCAAAQhAYPUEyuXyGQexuHLTTTctPmfHld1HXrFvaWng4gZL7jjz6Q/+4A9i//79S56N5JLya17B8EKbhZ5acwmfXUtf/vKX461vfWvt6XBW1+te97rFxyu509PTE7/1W7+VBKcbb7xxcVeXB1rEcyng1NRUysRaes7x8fHFbWt3LPJ98pOfTCWDS49lV9fv/u7v1jZLt3ap2c3W19e3+Py59l/cYIPfwYG1wSeY4UEAAhCAAAQgAAEIQAACEIAABFZLwKv6Wahxc2aThRiv5ldrDnb3SnsWYl5OwHruueeiVCq9RLzycSzQWMC60GYBzKHuS0Wsn/iJn4j/+l//a1Sr1XQYu6++9a1vpXK/Cz3u0u2c/fUjP/Ij8cu//Mvx1FNPpZfM4SMf+Ug4IN4int1mFqFqIe0uV7SY50wwi1C15vH9p//0n15SXunQeTvHlq7C+Pu///vxgQ98oLZruj3X/mdstIEf4MDawJPL0CAAAQhAAAIQgAAEIAABCEAAAmtB4P3vf39azc8r4h08eDD+/b//9ylkfOmxvRqhX1sekr50GwtU5wp49yqDP/qjPxou27uQ5tyo66+/PgWqW2CzcPYv/sW/SGV6t99+ezj76uTJk0lUu9gSQvfjV37lV9J4LSB1dnamUkeHujus3aWPbj//8z8fLh1saGhIQfIf/ehH43Of+1wcO3ZsMaz96quvTllXPs7SZgHsj/7oj5LzzMe38PWqV70qPKal7Vz7L91mI99vkOXtRb/dRh4pY4MABCAAAQhAAAIQgAAEIAABCEBgVQTsFHL2k0vnlrePfexjaYW9D3/4w8tfuqSPLRK51K4mJvlkDlH3aoG1lf7WqgMef0dHR/o52zEHBgbOGsh+tm3P9pzHYmfXy2WInW2/LDyHgJWFWWaMEIAABP7/9u4YRWEogKLo7D/7yV6yG8mAxTAWgu9WHkvBG3LKR/IlQIAAAQIECBAgQCAWuJ+sOo7jz9lY8SXlv0jg/2T6RTfvVgkQIECAAAECBAgQIECAAIHPBK7r+n2d7nkm1Gc1vybwWsAh7q9dfEuAAAECBAgQIECAAAECBAi8IXCe58/9L4X3weY+BCoBrxBWsroECBAgQIAAAQIECBAgQIAAAQITAU9gTRhFCBAgQIAAAQIECBAgQIAAAQIEKgEDViWrS4AAAQIECBAgQIAAAQIECBAgMBEwYE0YRQgQIECAAAECBAgQIECAAAECBCoBA1Ylq0uAAAECBAgQIECAAAECBAgQIDARMGBNGEUIECBAgAABAgQIECBAgAABAgQqAQNWJatLgAABAgQIECBAgAABAgQIECAwETBgTRhFCBAgQIAAAQIECBAgQIAAAQIEKgEDViWrS4AAAQIECBAgQIAAAQIECBAgMBEwYE0YRQgQIECAAAECBAgQIECAAAECBCoBA1Ylq0uAAAECBAgQIECAAAECBAgQIDARMGBNGEUIECBAgAABAgQIECBAgAABAgQqAQNWJatLgAABAgQIECBAgAABAgQIECAwETBgTRhFCBAgQIAAAQIECBAgQIAAAQIEKgEDViWrS4AAAQIECBAgQIAAAQIECBAgMBEwYE0YRQgQIECAAAECBAgQIECAAAECBCoBA1Ylq0uAAAECBAgQIECAAAECBAgQIDARMGBNGEUIECBAgAABAgQIECBAgAABAgQqAQNWJatLgAABAgQIECBAgAABAgQIECAwETBgTRhFCBAgQIAAAQIECBAgQIAAAQIEKgEDViWrS4AAAQIECBAgQIAAAQIECBAgMBEwYE0YRQgQIECAAAECBAgQIECAAAECBCoBA1Ylq0uAAAECBAgQIECAAAECBAgQIDARMGBNGEUIECBAgAABAgQIECBAgAABAgQqAQNWJatLgAABAgQIECBAgAABAgQIECAwETBgTRhFCBAgQIAAAQIECBAgQIAAAQIEKgEDViWrS4AAAQIECBAgQIAAAQIECBAgMBEwYE0YRQgQIECAAAECBAgQIECAAAECBCoBA1Ylq0uAAAECBAgQIECAAAECBAgQIDARMGBNGEUIECBAgAABAgQIECBAgAABAgQqAQNWJatLgAABAgQIECBAgAABAgQIECAwETBgTRhFCBAgQIAAAQIECBAgQIAAAQIEKgEDViWrS4AAAQIECBAgQIAAAQIECBAgMBEwYE0YRQgQIECAAAECBAgQIECAAAECBCoBA1Ylq0uAAAECBAgQIECAAAECBAgQIDARMGBNGEUIECBAgAABAgQIECBAgAABAgQqAQNWJatLgAABAgQIECBAgAABAgQIECAwETBgTRhFCBAgQIAAAQIECBAgQIAAAQIEKgEDViWrS4AAAQIECBAgQIAAAQIECBAgMBEwYE0YRQgQIECAAAECBAgQIECAAAECBCoBA1Ylq0uAAAECBAgQIECAAAECBAgQIDARMGBNGEUIECBAgAABAgQIECBAgAABAgQqAQNWJatLgAABAgQIECBAgAABAgQIECAwETBgTRhFCBAgQIAAAQIECBAgQIAAAQIEKgEDViWrS4AAAQIECBAgQIAAAQIECBAgMBEwYE0YRQgQIECAAAECBAgQIECAAAECBCoBA1Ylq0uAAAECBAgQIECAAAECBAgQIDARMGBNGEUIECBAgAABAgQIECBAgAABAgQqAQNWJatLgAABAgQIECBAgAABAgQIECAwETBgTRhFCBAgQIAAAQIECBAgQIAAAQIEKgEDViWrS4AAAQIECBAgQIAAAQIECBAgMBEwYE0YRQgQIECAAAECBAgQIECAAAECBCoBA1Ylq0uAAAECBAgQIECAAAECBAgQIDARMGBNGEUIECBAgAABAgQIECBAgAABAgQqAQNWJatLgAABAgQIECBAgAABAgQIECAwETBgTRhFCBAgQIAAAQIECBAgQIAAAQIEKgEDViWrS4AAAQIECBAgQIAAAQIECBAgMBEwYE0YRQgQIECAAAECBAgQIECAAAECBCoBA1Ylq0uAAAECBAgQIECAAAECBAgQIDAReACvROb2YYdurwAAAABJRU5ErkJggg==\n"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%%R -i embedding --width 1200 --height 800 -r 140 --bg #F5F5F5\n",
    "library(hrbrthemes)\n",
    "library(ggplot2)\n",
    "library(fields)\n",
    "embedding$density <- fields::interp.surface(\n",
    "  MASS::kde2d(embedding$x, embedding$y), embedding[,c(\"x\",\"y\")])\n",
    "\n",
    "p <- ggplot(embedding, aes(x=embedding$x, y=embedding$y,alpha = 1/density))+\n",
    "\n",
    " guides(alpha=FALSE)+\n",
    "\n",
    "geom_point(color=\"#3366cc\", pch=16,cex=1.2)+ theme_ipsum_rc()+\n",
    "labs(x=\"\", y=\"\",\n",
    "       title=\"The 2d-reduction by UMAP\",\n",
    "       subtitle=\"...based on the code by McInnes, Healy (2018) and Noichl (2018)\",\n",
    "       caption=\"by Anton Savov\")+\n",
    "theme(panel.grid.major = element_line(colour = \"lightgrey\"),panel.grid.minor = element_blank())+\n",
    "\n",
    "#comment out the next two lines to see the whole space with all the outliers\n",
    "scale_x_continuous(limits=c(-25,25),labels=NULL, breaks=seq(-25,25,length=16))+\n",
    "scale_y_continuous(limits=c(-15,15),labels=NULL, breaks=seq(-15,15,length=10))+\n",
    "\n",
    "NULL\n",
    "p"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Clustering\n",
    "For the clustering we use [HDBSCAN](https://github.com/scikit-learn-contrib/hdbscan).\n",
    "The `clustersize` variable needs to be adjusted based on the number of papers in the data set. For me and my collection of 12,000 records a number between 100 and 200 worked well.\n",
    "\n",
    "It is important to note that the process of embedding into two dimensions with UMAP is stochastic, i.e. non-deterministic. In plain text that means that each time we run it it will produce slightly different results even though we use the same data. This also has influence on the clustering. In one of the runs I did a cluster size of 100 split the map into 24 clusters, in the next run the same data and cluster size resulted in 17 clusters. So bare that in mind when judging the final result presented here and also if you use this method to produce your own maps."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 144,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "17\n"
     ]
    }
   ],
   "source": [
    "try:\n",
    "    drc = drc.drop('cluster',axis=1)\n",
    "except KeyError:\n",
    "    pass\n",
    "\n",
    "import hdbscan\n",
    "\n",
    "#(min_cluster_size=500, min_samples=30, gen_min_span_tree=True)\n",
    "#clusterer = hdbscan.HDBSCAN(min_cluster_size=455, min_samples=35, gen_min_span_tree=True)\n",
    "\n",
    "#adjust this number based on number of papers in your data set\n",
    "clustersize = 140\n",
    "\n",
    "clusterer = hdbscan.HDBSCAN(min_cluster_size=clustersize, min_samples=50, gen_min_span_tree=True)\n",
    "clusterer.fit(embedding)\n",
    "XCLUST = clusterer.labels_\n",
    "clusternum = len(set( clusterer.labels_))-1\n",
    "\n",
    "\n",
    "dfclust = pd.DataFrame(XCLUST)\n",
    "dfclust.columns = ['cluster']\n",
    "\n",
    "print(clusternum)\n",
    "### Let's play a little sound when we're done:\n",
    "#import winsound\n",
    "#winsound.Beep(550,300)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now lets plot the map and the clusters in `ggplot`:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 280,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%%R -i embedding,dfclust -o myNewColors --width 1200 --height 800 -r 140 --bg #F8F4E9\n",
    "\n",
    "# --bg #231f20\n",
    "\n",
    "library(hrbrthemes)\n",
    "library(ggplot2)\n",
    "library(fields)\n",
    "library(plyr)\n",
    "options(warn=0)# 0 zum anschalten\n",
    "\n",
    "#Get the cluster means:\n",
    "means <- aggregate(embedding[,c(\"x\",\"y\")], list(dfclust$cluster), median)\n",
    "means <- data.frame(means) \n",
    "n=nrow(means)\n",
    "means <- means[-1,]\n",
    "\n",
    "#good place to try get consistent colors is http://colorbrewer2.org/\n",
    "#Make the colors: \n",
    "mycolors <- c(\"#6abc70\", \"#0abcce\", \"#f9f4b8\", \"#ffe800\", \"#fbab19\", \"#f68121\", \"#ed1c24\", \"#ca205f\") \n",
    "\n",
    "pal <- colorRampPalette(sample(mycolors))\n",
    "s <- n-1\n",
    "myGray <- c('#aaaaaa')\n",
    "myNewColors <- sample(pal(s))\n",
    "myPal <- append(myGray,myNewColors)\n",
    "\n",
    "\n",
    "#get density, to avoid overplotting\n",
    "embedding$density <- fields::interp.surface(\n",
    "  MASS::kde2d(embedding$x, embedding$y), embedding[,c(\"x\",\"y\")])\n",
    "\n",
    "p <- ggplot(embedding, aes(x=embedding$x, y=embedding$y, color= as.factor(dfclust$cluster), alpha = 1/density))+\n",
    "geom_point(pch=20,cex=1.6)+ \n",
    "theme_ipsum_rc()+\n",
    "scale_color_manual(values = myPal) +\n",
    " guides(alpha=FALSE, color=FALSE)+\n",
    "geom_point(data=means, aes(x=means$x, y=means$y), color= myNewColors, alpha = 1,size =6)+\n",
    "#annotate(\"text\", x = means[,c(\"x\")], y = means[,c(\"y\")], label = means[,c(\"Group.1\")], color=\"white\", fontface=\"bold\",  size=3.2, parse = TRUE, hjust=0.5)+\n",
    "annotate(\"text\", x = means[,c(\"x\")], y = means[,c(\"y\")], label = means[,c(\"Group.1\")], color=\"white\", fontface=\"bold\",  size=3.2, parse = TRUE, hjust=0.5)+\n",
    "labs(x=\"\", y=\"\",\n",
    "       title=\"The clusters found by hdbscan...\",\n",
    "       subtitle=\"a density-based clustering algorithm. Embedded with UMAP in two dimensions...\")+\n",
    "theme(panel.grid.major = element_line(colour = \"lightgrey\", size=0),panel.grid.minor = element_blank())+\n",
    "#theme(panel.grid.major = element_blank(),panel.grid.minor = element_blank())\n",
    "\n",
    "#comment out the next two lines to see the whole space with all the outliers\n",
    "scale_x_continuous(limits=c(-25,25),labels=NULL, breaks=seq(-25,25,length=16))+\n",
    "scale_y_continuous(limits=c(-15,15),labels=NULL, breaks=seq(-15,15,length=10))+\n",
    "\n",
    "NULL\n",
    "\n",
    "p\n",
    "\n",
    "#pdf(\"ClusteringUMap.pdf\", width = 12, height = 12) # Open a new pdf file\n",
    "#print(p) #write the plot to the PDF\n",
    "#dev.off() #close the PDF"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Tree of cluster similarity\n",
    "\n",
    "Now that we have the clusters we can look at which clusters are more similar to each other by taking a look at when did the algorithm found it necessary to split a group into two.\n",
    "In the chart below, the sooner a cluster broke off of the main trunk, the more disconnected are the articles it contains from the rest of the research body.\n",
    "\n",
    "The clusters are numbered and later we will analyse the content of each cluster so you can scroll down to the labels to check what body of work is in each cluster.\n",
    "\n",
    "My observation is that \"Shape Grammars\" are usually the first cluster to split from the bulk, so probably it means they are the most distinguishable in terms of the features we use (cited authors and cited works) from every other research topic in digital design in architecture.\n",
    "\n",
    "In general it is interesting to observe in this tree where things split from each other. If we take shape grammars out then what is the next most distinguishable field of research?\n",
    "\n",
    "### Colours\n",
    "Next we will generate colours for the cluster from a gradient pallette using the cluster id or the lambda value of when it split off from the main trunk as a sampling variable."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 231,
   "metadata": {},
   "outputs": [],
   "source": [
    "#define cluster colours by split lambda value\n",
    "tree = clusterer.condensed_tree_.to_pandas()\n",
    "cluster_tree = tree[tree.child_size > 1]\n",
    "#display(cluster_tree)\n",
    "from matplotlib import cm\n",
    "from matplotlib.colors import to_hex\n",
    "\n",
    "#the reference for the built in linear color maps in matplotlib is at:\n",
    "#  https://matplotlib.org/3.1.1/gallery/color/colormap_reference.html \n",
    "#cluster_gradient = cm.cool\n",
    "#cluster_gradient = cm.rainbow\n",
    "#cluster_gradient = cm.gist_rainbow\n",
    "#cluster_gradient = cm.jet\n",
    "\n",
    "#cluster_gradient = matplotlib.colors.LinearSegmentedColormap.from_list(\"\", [\"#4D3E71\",\"#7D4495\",\"#AB5596\",\"#DA6688\",\"#F88F79\",\"#FCC696\"])\n",
    "cluster_gradient = matplotlib.colors.LinearSegmentedColormap.from_list(\n",
    "    \"\",\n",
    "    [\"#FCC696\",\"#F88F79\",\"#DA6688\",\"#AB5596\",\"#7D4495\",\"#4D3E71\"]\n",
    ")\n",
    "#cluster_gradient = matplotlib.colors.LinearSegmentedColormap.from_list(\"\", [\"#009392\",\"#39b185\",\"#9ccb86\",\"#e9e29c\",\"#eeb479\",\"#e88471\",\"#cf597e\"])\n",
    "#cluster_gradient = cm.hsv\n",
    "\n",
    "#get the node ids of the clusters\n",
    "chosen_clusters = clusterer.condensed_tree_._select_clusters()\n",
    "\n",
    "#get the condensed cluster tree\n",
    "tree = clusterer.condensed_tree_.to_numpy()\n",
    "cluster_tree = [x for x in tree if x[3]>1]\n",
    "#display(cluster_tree)\n",
    "\n",
    "lambda_vals = [x[2] for x in cluster_tree if x[1] in chosen_clusters]\n",
    "#display(lambda_vals)\n",
    "\n",
    "from scipy.interpolate import interp1d\n",
    "m = interp1d([min(lambda_vals),max(lambda_vals)],[0,1])\n",
    "#print(m(lambda_vals[1]))\n",
    "#print(to_hex(cluster_gradient(.5)))\n",
    "my_cluster_colours = [to_hex(cluster_gradient(m(x))) for x in lambda_vals]\n",
    "#display(my_cluster_colours)\n",
    "\n",
    "from rpy2 import robjects\n",
    "\n",
    "my_cluster_colours_R = robjects.vectors.StrVector(my_cluster_colours)\n",
    "#print(type(my_cluster_colours_R))\n",
    "#display(my_cluster_colours_R)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['#fcc696',\n",
       " '#fbb58d',\n",
       " '#f9a384',\n",
       " '#f8927b',\n",
       " '#f0857d',\n",
       " '#e77882',\n",
       " '#de6b86',\n",
       " '#d1638b',\n",
       " '#c25d8f',\n",
       " '#b35894',\n",
       " '#a55396',\n",
       " '#964d96',\n",
       " '#884895',\n",
       " '#794492',\n",
       " '#6a4287',\n",
       " '#5b407c',\n",
       " '#4d3e71']"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#or, alterntatively, define colors by cluster numbers\n",
    "from matplotlib import cm\n",
    "from matplotlib.colors import to_hex, LinearSegmentedColormap\n",
    "\n",
    "#cluster_gradient = matplotlib.colors.LinearSegmentedColormap.from_list(\"\", [\"#4D3E71\",\"#7D4495\",\"#AB5596\",\"#DA6688\",\"#F88F79\",\"#FCC696\"])\n",
    "cluster_gradient = LinearSegmentedColormap.from_list(\n",
    "    \"\",\n",
    "    [\"#FCC696\",\"#F88F79\",\"#DA6688\",\"#AB5596\",\"#7D4495\",\"#4D3E71\"]\n",
    ")\n",
    "#cluster_gradient = matplotlib.colors.LinearSegmentedColormap.from_list(\"\", [\"#009392\",\"#39b185\",\"#9ccb86\",\"#e9e29c\",\"#eeb479\",\"#e88471\",\"#cf597e\"])\n",
    "#cluster_gradient = cm.hsv\n",
    "\n",
    "eqspace = np.linspace(0.0, 1.0, num=17)\n",
    "my_cluster_colours_2 = [to_hex(cluster_gradient(x)) for x in eqspace]\n",
    "display(my_cluster_colours_2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Next let's use `SankeyWidget`to plot the cluster tree:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 232,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "070916ca29a8466da774fe5170a58f99",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "SankeyWidget(layout=Layout(height='300', width='1920'), links=[{'source': 9779, 'target': 9780, 'value': 415, …"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from ipysankeywidget import SankeyWidget\n",
    "from ipywidgets import Layout\n",
    "from IPython.display import (\n",
    "    Image,\n",
    "    SVG\n",
    ")\n",
    "\n",
    "parent_nodes = [x[0] for x in cluster_tree]\n",
    "\n",
    "links = []\n",
    "cluster_nodes =[]\n",
    "nodes = []\n",
    "removed_parent_nodes = []\n",
    "\n",
    "first_node = min(parent_nodes)\n",
    "node = {'id': first_node, 'title': 'All articles' }\n",
    "nodes.append(node)\n",
    "\n",
    "for edge in cluster_tree:\n",
    "    if edge[0] in chosen_clusters or edge[0] in removed_parent_nodes:\n",
    "         #filter out leaves that didn't get selected as a cluster\n",
    "        removed_parent_nodes.append(edge[1])\n",
    "    else:\n",
    "        if edge[1] in chosen_clusters:\n",
    "            #this is a cluster leaf so set label, color, etc\n",
    "            cluster_id = chosen_clusters.index(edge[1])\n",
    "            cluster_label = \"cluster \" + str(cluster_id)\n",
    "            color = my_cluster_colours[cluster_id]\n",
    "            cluster_nodes.append(edge[1])\n",
    "            node = {'id': edge[1], 'title': cluster_label, 'style': 'process' }\n",
    "        else:\n",
    "            #this is a branch so use default color and remove label\n",
    "            color = '#16457D' #\"#bfc4c6\"\n",
    "            node = {'id': edge[1], 'title': '' }\n",
    "        nodes.append(node)\n",
    "        link = {\n",
    "            'source':edge[0],\n",
    "            'target':edge[1],\n",
    "            'value':edge[3],\n",
    "            'color':color\n",
    "        }\n",
    "        links.append(link)\n",
    "    \n",
    "#display(links)\n",
    "#display(cluster_nodes)\n",
    "\n",
    "#use the rank_sets to place all cluster leafs on the same line\n",
    "rank_sets = [\n",
    "    { 'type': 'same', 'nodes': cluster_nodes }\n",
    "]\n",
    "\n",
    "#use the order to force order by splitting event\n",
    "#to use pass order=order in the arguments of SankeyWidget\n",
    "#It creates kind of a wavey chart so I decided not to use it\n",
    "'''\n",
    "sorted_nodes = []\n",
    "for edge in cluster_tree:\n",
    "    for i in [0,1]:\n",
    "        if not edge[i] in removed_parent_nodes and not edge[i] in sorted_nodes:\n",
    "            sorted_nodes.append(edge[i])\n",
    "order = []\n",
    "for node in sorted(sorted_nodes):\n",
    "    order.append([node])\n",
    "'''\n",
    "\n",
    "#sankey = SankeyWidget(links=links, rank_sets=rank_sets, nodes=nodes, layout=Layout(width=\"2400\", height=\"400\")).auto_save_png('test.png')\n",
    "sankey = SankeyWidget(links=links, nodes=nodes, layout = Layout(width=\"1920\", height=\"300\")).auto_save_svg('exports/cluster_tree.svg')\n",
    "sankey\n",
    "\n",
    "#SVG(sankey.svg)\n",
    "#sankey.save_svg('test.svg')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here is the result of the last cell:\n",
    "![Cluster Tree](images/cluster_tree.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## What is each cluster about?\n",
    "Now, let's figure out what body or work does each cluster contain. Lets look at the common words, the cited works and authors for each cluster. \n",
    "\n",
    "### The most common words and phrases\n",
    "We can analyze the titles and the abstract of the articles in a cluster and derive the most common words and phrases. Since in the field there are established threee-word-terms such as \"Building Information Modeling\" it makes sense to look for the most common \n",
    "\n",
    "And since some article records don't have the abstract info in the data we downloaded from WOS we will do this analysis separate for the titles and the abstract. Plus the abstract have much more words than the titles so their content will drown whatever the titles are about.\n",
    "\n",
    "But before we do any text analysis lets unify the spelling differences between US and UK English such as \"modeling\" and \"modelling\". And we will also remove the most common words such as \"paper\", as well as the terms that are common accross all clusters such as \"design\"."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "metadata": {},
   "outputs": [],
   "source": [
    "drc = pd.concat([drc, dfclust],axis=1)\n",
    "drc = drc.dropna(subset=['cluster'])\n",
    "drc = pd.concat([drc, embedding],axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [],
   "source": [
    "#load up the UK to US spelling dictionary\n",
    "#the list of words is taken from http://www.tysto.com/uk-us-spelling-list.html\n",
    "#import random\n",
    "uk2us = {}\n",
    "with open('data/uk-to-us-dict.csv', 'r') as csvfile:\n",
    "    spamreader = csv.reader(csvfile, delimiter=';')\n",
    "    for row in spamreader:\n",
    "        ukword = row[0].replace(\" \", \"\")\n",
    "        usword = row[1].replace(\" \", \"\")\n",
    "        uk2us[ukword] = usword\n",
    "#print(uk2us[random.choice(list(uk2us.keys()))])\n",
    "\n",
    "#this function will replace words that are keys in mydict with the corresponding value\n",
    "def replace_all(text, mydict):\n",
    "    for gb, us in mydict.items():\n",
    "        #text = text.replace(us, gb)\n",
    "        text = re.sub(r\"\\b%s\\b\" % re.escape(gb) , us, text)\n",
    "    return text"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {},
   "outputs": [],
   "source": [
    "#kick out common abstract words of no importance\n",
    "removeWords = ['|','paper','argue','account','theory',\n",
    "                'elsevier','reserved','rights',\n",
    "                'used','use','new','using',\n",
    "                'copyright', 'published']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {},
   "outputs": [],
   "source": [
    "#builds the text bag for each cluster for a given tag\n",
    "def buildTextBags(drcval, replacedict, removelist, tag, clean=True):\n",
    "    fullstrsl = []\n",
    "    \n",
    "\n",
    "    #build a map of the words to remove for later\n",
    "    big_regex = re.compile(r'\\b%s\\b' % r'\\b|\\b'.join(map(re.escape, removelist)))\n",
    "    for x in range(0,clusternum):\n",
    "        textBag = list(drcval.loc[drcval['cluster'] == x][tag])\n",
    "        #titles = list(drc.loc[drc['cluster'] == x]['title'])\n",
    "        #abstracts = abstracts + titles\n",
    "        #abstracts = \";\".join(str(x) for x in abstracts).replace('|',' ').replace('paper',' ').replace('argue',' ').replace('account',' ').replace('theory',' ').replace('elsevier',' ').replace('reserved',' ').replace('rights',' ').replace('used',' ').replace('use',' ').replace('new',' ').replace('using',' ').replace('process',' ').replace('approach',' ') #kick out common abstract words of no importance\n",
    "        textBag = \";\".join(str(x) for x in textBag) \n",
    "        if clean:\n",
    "            textBag = big_regex.sub(\" \", textBag)#kick out common abstract words of no importance\n",
    "        #use the word replacing code here https://bytes.com/topic/python/answers/795690-replacing-string-words-list#post3162180\n",
    "        textBag = replace_all(textBag, replacedict) #this slows things down\n",
    "        fullstrsl.append(textBag)\n",
    "    #print(fullstrsl[0][:50])\n",
    "    return fullstrsl\n",
    "\n",
    "#build the bags - kind of slow for the abstracts \n",
    "clusterTextBagsTitles = buildTextBags(drc, uk2us, removeWords, 'title')\n",
    "clusterTextBagsAbstracts = buildTextBags(drc, uk2us, removeWords, 'abstract')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {},
   "outputs": [],
   "source": [
    "#check if a string searchfor exists in 'searchin' only as a whole word/phrase\n",
    "def string_found(searchfor, searchin):\n",
    "    return re.search(r'\\b%s\\b' % (re.escape(searchfor)), searchin) is not None\n",
    "\n",
    "#if a phrase is found in a longer phrase reduce its score with the score of hte longer phrase\n",
    "#exclude if true the word score is set to 0 if exclude is false then the longer word score is subtracted from short word score\n",
    "def accountScores(longphrasescores, shortphrasescores, exclude=True):\n",
    "    newscores =[]\n",
    "    for phrase in shortphrasescores:\n",
    "        newphraseScore = [phrase[0],phrase[1]]\n",
    "        for longphrase in longphrasescores:\n",
    "            #if phrase is found in longphrase subtract scores\n",
    "            #pat = re.compile(r\"\\b%s\\b\" % re.escape(phrase[0]))\n",
    "            #print(phrase[0])\n",
    "            \n",
    "            if string_found(phrase[0], longphrase[0]):\n",
    "                if exclude:\n",
    "                    newphraseScore[1] = 0\n",
    "                else:                      \n",
    "                    newphraseScore[1] = phrase[1]-longphrase[1]\n",
    "        newscores.append(newphraseScore)\n",
    "    return newscores\n",
    "\n",
    "#this def below is not needed anymore\n",
    "def accountScoreClusters(longscores, shortscores):\n",
    "    counter =0\n",
    "    newclusterscores =[]\n",
    "    for cluster in shortscores:\n",
    "        newclusterscores.append(accountScores(longscores[counter],cluster))\n",
    "        counter +=1\n",
    "    return newclusterscores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "metadata": {},
   "outputs": [],
   "source": [
    "#tag is which tag('abstract' or 'title') to analyze for\n",
    "#phrase is the phrase length in words. use 2 for bigrams;\n",
    "#n is how many words to output\n",
    "#treshold is the minimum number of occurences needed to include the word\n",
    "#clean if True will skip the words in the removeWords list\n",
    "#longphrasescores is a list of pairs\n",
    "#           first item is scorelist for already ranked words/phrases\n",
    "#           second item if true the word score is set to 0 if exclude is false then the longer word score is subtracted from short word score\n",
    "#longphrasescores is used for example to reduce the score of the bigram \"building information\" with the score of the trigram \"building information modeling\", i.e. to avoid double counting\n",
    "def rankWords(textBag,phrase=1, n=10,treshold=10, longphrasescores=None):\n",
    "    \n",
    "    \n",
    "    vec = CountVectorizer( stop_words='english', ngram_range=(phrase, phrase))#Choose CountVectorizer for the most common words in the cluster, TfidfVectorizer for the words with the greatest differentiation value.\n",
    "    X = vec.fit_transform(textBag)\n",
    "    #print(pd.DataFrame(X.toarray(), columns=vec.get_feature_names())) #To look into the vectors. Beware, can take a bit of RAM\n",
    "    \n",
    "    \n",
    "    clusterfeatures = pd.DataFrame(X.toarray(), columns=vec.get_feature_names())\n",
    "    fullscore = []\n",
    "    for x in range(0,clusternum):\n",
    "        scores = zip(vec.get_feature_names(), np.asarray(X[x,:].sum(axis=0)).ravel())\n",
    "        \n",
    "        accountedScores = sorted(scores, key=lambda x: x[1], reverse=True)\n",
    "        accountedScores = accountedScores[0:60]\n",
    "        #subtract scores if phrase is found in longphrasescores\n",
    "        if longphrasescores is not None:            \n",
    "            for longphrasescore in longphrasescores:\n",
    "                accountedScores = accountScores(longphrasescore[0][x], accountedScores,longphrasescore[1])\n",
    "        #sort\n",
    "        sorted_scores = sorted(accountedScores, key=lambda x: x[1], reverse=True)\n",
    "        \n",
    "        myscores = sorted_scores[0:n]\n",
    "        #print(myscores)\n",
    "        #print(\"___________________\")\n",
    "        scorelist = []\n",
    "        counter = 0\n",
    "        for s in myscores:\n",
    "            #print(s)\n",
    "            if s[1]>treshold or counter == 0:\n",
    "                #newWord = s[0] + ' ('+str(s[1]) + ')' # do the word and the how often it comes\n",
    "                newWord = [s[0],s[1]] # do the word and the how often it comes\n",
    "                scorelist.append(newWord)\n",
    "            #else:\n",
    "            #    newWord = 'x' # skip the word if it appears less than 10 times\n",
    "            \n",
    "            counter += 1\n",
    "        fullscore.append(scorelist)\n",
    "    return fullscore\n",
    "\n",
    "#adds to dataframes of scores for one cluster together\n",
    "def addClusterScores(score1,score2):\n",
    "    newscore = []\n",
    "    for clusterid in range(0,len(score1)):\n",
    "        clusterscore =[]\n",
    "        for item in score1[clusterid]:\n",
    "            clusterscore.append(item)\n",
    "        for item in score2[clusterid]:\n",
    "            clusterscore.append(item)   \n",
    "        newscore.append(clusterscore)\n",
    "    return newscore"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TRIGRAMS ABSTRACTS\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>10</th>\n",
       "      <th>11</th>\n",
       "      <th>12</th>\n",
       "      <th>13</th>\n",
       "      <th>14</th>\n",
       "      <th>15</th>\n",
       "      <th>16</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>[shape grammar interpreter, 10]</td>\n",
       "      <td>[point cloud data, 17]</td>\n",
       "      <td>[space syntax analysis, 8]</td>\n",
       "      <td>[planning support systems, 10]</td>\n",
       "      <td>[cellular automata ca, 17]</td>\n",
       "      <td>[design digital fabrication, 10]</td>\n",
       "      <td>[building information modeling, 5]</td>\n",
       "      <td>[finite element analysis, 9]</td>\n",
       "      <td>[geometric constraint solving, 14]</td>\n",
       "      <td>[building information modeling, 106]</td>\n",
       "      <td>[finite element analysis, 10]</td>\n",
       "      <td>[architectural design process, 11]</td>\n",
       "      <td>[building information modeling, 64]</td>\n",
       "      <td>[human behavior simulation, 6]</td>\n",
       "      <td>[building information modeling, 8]</td>\n",
       "      <td>[analog digital notational, 2]</td>\n",
       "      <td>[creative conceptual design, 9]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>None</td>\n",
       "      <td>[radio frequency identification, 16]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[information modeling bim, 76]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[information modeling bim, 47]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>None</td>\n",
       "      <td>[building information models, 13]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[industry foundation classes, 30]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[tool path generation, 39]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>None</td>\n",
       "      <td>[precast concrete elements, 12]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[architecture engineering construction, 24]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[support vector machine, 29]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>None</td>\n",
       "      <td>[building information modeling, 11]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[building information models, 22]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[computer aided design, 18]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[foundation classes ifc, 21]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[genetic algorithm ga, 17]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[semantic web technologies, 14]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[particle swarm optimization, 16]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[building information model, 11]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[finite element analysis, 15]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[bv building information, 11]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[experimental results demonstrate, 14]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[engineering construction aec, 11]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[finite element method, 14]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                0                                     1   \\\n",
       "0  [shape grammar interpreter, 10]                [point cloud data, 17]   \n",
       "1                             None  [radio frequency identification, 16]   \n",
       "2                             None     [building information models, 13]   \n",
       "3                             None       [precast concrete elements, 12]   \n",
       "4                             None   [building information modeling, 11]   \n",
       "5                             None                                  None   \n",
       "6                             None                                  None   \n",
       "7                             None                                  None   \n",
       "8                             None                                  None   \n",
       "9                             None                                  None   \n",
       "\n",
       "                           2                               3   \\\n",
       "0  [space syntax analysis, 8]  [planning support systems, 10]   \n",
       "1                        None                            None   \n",
       "2                        None                            None   \n",
       "3                        None                            None   \n",
       "4                        None                            None   \n",
       "5                        None                            None   \n",
       "6                        None                            None   \n",
       "7                        None                            None   \n",
       "8                        None                            None   \n",
       "9                        None                            None   \n",
       "\n",
       "                           4                                 5   \\\n",
       "0  [cellular automata ca, 17]  [design digital fabrication, 10]   \n",
       "1                        None                              None   \n",
       "2                        None                              None   \n",
       "3                        None                              None   \n",
       "4                        None                              None   \n",
       "5                        None                              None   \n",
       "6                        None                              None   \n",
       "7                        None                              None   \n",
       "8                        None                              None   \n",
       "9                        None                              None   \n",
       "\n",
       "                                   6                             7   \\\n",
       "0  [building information modeling, 5]  [finite element analysis, 9]   \n",
       "1                                None                          None   \n",
       "2                                None                          None   \n",
       "3                                None                          None   \n",
       "4                                None                          None   \n",
       "5                                None                          None   \n",
       "6                                None                          None   \n",
       "7                                None                          None   \n",
       "8                                None                          None   \n",
       "9                                None                          None   \n",
       "\n",
       "                                   8   \\\n",
       "0  [geometric constraint solving, 14]   \n",
       "1                                None   \n",
       "2                                None   \n",
       "3                                None   \n",
       "4                                None   \n",
       "5                                None   \n",
       "6                                None   \n",
       "7                                None   \n",
       "8                                None   \n",
       "9                                None   \n",
       "\n",
       "                                            9                              10  \\\n",
       "0         [building information modeling, 106]  [finite element analysis, 10]   \n",
       "1               [information modeling bim, 76]                           None   \n",
       "2            [industry foundation classes, 30]                           None   \n",
       "3  [architecture engineering construction, 24]                           None   \n",
       "4            [building information models, 22]                           None   \n",
       "5                 [foundation classes ifc, 21]                           None   \n",
       "6              [semantic web technologies, 14]                           None   \n",
       "7             [building information model, 11]                           None   \n",
       "8                [bv building information, 11]                           None   \n",
       "9           [engineering construction aec, 11]                           None   \n",
       "\n",
       "                                   11                                      12  \\\n",
       "0  [architectural design process, 11]     [building information modeling, 64]   \n",
       "1                                None          [information modeling bim, 47]   \n",
       "2                                None              [tool path generation, 39]   \n",
       "3                                None            [support vector machine, 29]   \n",
       "4                                None             [computer aided design, 18]   \n",
       "5                                None              [genetic algorithm ga, 17]   \n",
       "6                                None       [particle swarm optimization, 16]   \n",
       "7                                None           [finite element analysis, 15]   \n",
       "8                                None  [experimental results demonstrate, 14]   \n",
       "9                                None             [finite element method, 14]   \n",
       "\n",
       "                               13                                  14  \\\n",
       "0  [human behavior simulation, 6]  [building information modeling, 8]   \n",
       "1                            None                                None   \n",
       "2                            None                                None   \n",
       "3                            None                                None   \n",
       "4                            None                                None   \n",
       "5                            None                                None   \n",
       "6                            None                                None   \n",
       "7                            None                                None   \n",
       "8                            None                                None   \n",
       "9                            None                                None   \n",
       "\n",
       "                               15                               16  \n",
       "0  [analog digital notational, 2]  [creative conceptual design, 9]  \n",
       "1                            None                             None  \n",
       "2                            None                             None  \n",
       "3                            None                             None  \n",
       "4                            None                             None  \n",
       "5                            None                             None  \n",
       "6                            None                             None  \n",
       "7                            None                             None  \n",
       "8                            None                             None  \n",
       "9                            None                             None  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TRIGRAMS TITLES\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>10</th>\n",
       "      <th>11</th>\n",
       "      <th>12</th>\n",
       "      <th>13</th>\n",
       "      <th>14</th>\n",
       "      <th>15</th>\n",
       "      <th>16</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>[early stages design, 3]</td>\n",
       "      <td>[3d laser scanning, 4]</td>\n",
       "      <td>[urban street networks, 3]</td>\n",
       "      <td>[positive theory planning, 3]</td>\n",
       "      <td>[agent based modeling, 6]</td>\n",
       "      <td>[building information modelling, 3]</td>\n",
       "      <td>[design process planning, 2]</td>\n",
       "      <td>[algorithm automatic 2d, 2]</td>\n",
       "      <td>[geometric constraint systems, 6]</td>\n",
       "      <td>[building information modelling, 5]</td>\n",
       "      <td>[free form surfaces, 7]</td>\n",
       "      <td>[architectural design studio, 2]</td>\n",
       "      <td>[sculptured surface machining, 9]</td>\n",
       "      <td>[classicism new synthesis, 2]</td>\n",
       "      <td>[land use planning, 8]</td>\n",
       "      <td>[art net london, 2]</td>\n",
       "      <td>[case based design, 8]</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                         0                       1   \\\n",
       "0  [early stages design, 3]  [3d laser scanning, 4]   \n",
       "\n",
       "                           2                              3   \\\n",
       "0  [urban street networks, 3]  [positive theory planning, 3]   \n",
       "\n",
       "                          4                                    5   \\\n",
       "0  [agent based modeling, 6]  [building information modelling, 3]   \n",
       "\n",
       "                             6                            7   \\\n",
       "0  [design process planning, 2]  [algorithm automatic 2d, 2]   \n",
       "\n",
       "                                  8                                    9   \\\n",
       "0  [geometric constraint systems, 6]  [building information modelling, 5]   \n",
       "\n",
       "                        10                                11  \\\n",
       "0  [free form surfaces, 7]  [architectural design studio, 2]   \n",
       "\n",
       "                                  12                             13  \\\n",
       "0  [sculptured surface machining, 9]  [classicism new synthesis, 2]   \n",
       "\n",
       "                       14                   15                      16  \n",
       "0  [land use planning, 8]  [art net london, 2]  [case based design, 8]  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TRIGRAMS COMBINED\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>10</th>\n",
       "      <th>11</th>\n",
       "      <th>12</th>\n",
       "      <th>13</th>\n",
       "      <th>14</th>\n",
       "      <th>15</th>\n",
       "      <th>16</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>[shape grammar interpreter, 10]</td>\n",
       "      <td>[point cloud data, 17]</td>\n",
       "      <td>[space syntax analysis, 8]</td>\n",
       "      <td>[planning support systems, 10]</td>\n",
       "      <td>[cellular automata ca, 17]</td>\n",
       "      <td>[design digital fabrication, 10]</td>\n",
       "      <td>[building information modeling, 5]</td>\n",
       "      <td>[finite element analysis, 9]</td>\n",
       "      <td>[geometric constraint solving, 14]</td>\n",
       "      <td>[building information modeling, 106]</td>\n",
       "      <td>[finite element analysis, 10]</td>\n",
       "      <td>[architectural design process, 11]</td>\n",
       "      <td>[building information modeling, 64]</td>\n",
       "      <td>[human behavior simulation, 6]</td>\n",
       "      <td>[building information modeling, 8]</td>\n",
       "      <td>[analog digital notational, 2]</td>\n",
       "      <td>[creative conceptual design, 9]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>[early stages design, 3]</td>\n",
       "      <td>[radio frequency identification, 16]</td>\n",
       "      <td>[urban street networks, 3]</td>\n",
       "      <td>[positive theory planning, 3]</td>\n",
       "      <td>[agent based modeling, 6]</td>\n",
       "      <td>[building information modelling, 3]</td>\n",
       "      <td>[design process planning, 2]</td>\n",
       "      <td>[algorithm automatic 2d, 2]</td>\n",
       "      <td>[geometric constraint systems, 6]</td>\n",
       "      <td>[information modeling bim, 76]</td>\n",
       "      <td>[free form surfaces, 7]</td>\n",
       "      <td>[architectural design studio, 2]</td>\n",
       "      <td>[information modeling bim, 47]</td>\n",
       "      <td>[classicism new synthesis, 2]</td>\n",
       "      <td>[land use planning, 8]</td>\n",
       "      <td>[art net london, 2]</td>\n",
       "      <td>[case based design, 8]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>None</td>\n",
       "      <td>[building information models, 13]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[industry foundation classes, 30]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[tool path generation, 39]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>None</td>\n",
       "      <td>[precast concrete elements, 12]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[architecture engineering construction, 24]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[support vector machine, 29]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>None</td>\n",
       "      <td>[building information modeling, 11]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[building information models, 22]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[computer aided design, 18]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>None</td>\n",
       "      <td>[3d laser scanning, 4]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[foundation classes ifc, 21]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[genetic algorithm ga, 17]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[semantic web technologies, 14]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[particle swarm optimization, 16]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[building information model, 11]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[finite element analysis, 15]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[bv building information, 11]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[experimental results demonstrate, 14]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[engineering construction aec, 11]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[finite element method, 14]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[building information modelling, 5]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[sculptured surface machining, 9]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                 0                                     1   \\\n",
       "0   [shape grammar interpreter, 10]                [point cloud data, 17]   \n",
       "1          [early stages design, 3]  [radio frequency identification, 16]   \n",
       "2                              None     [building information models, 13]   \n",
       "3                              None       [precast concrete elements, 12]   \n",
       "4                              None   [building information modeling, 11]   \n",
       "5                              None                [3d laser scanning, 4]   \n",
       "6                              None                                  None   \n",
       "7                              None                                  None   \n",
       "8                              None                                  None   \n",
       "9                              None                                  None   \n",
       "10                             None                                  None   \n",
       "\n",
       "                            2                               3   \\\n",
       "0   [space syntax analysis, 8]  [planning support systems, 10]   \n",
       "1   [urban street networks, 3]   [positive theory planning, 3]   \n",
       "2                         None                            None   \n",
       "3                         None                            None   \n",
       "4                         None                            None   \n",
       "5                         None                            None   \n",
       "6                         None                            None   \n",
       "7                         None                            None   \n",
       "8                         None                            None   \n",
       "9                         None                            None   \n",
       "10                        None                            None   \n",
       "\n",
       "                            4                                    5   \\\n",
       "0   [cellular automata ca, 17]     [design digital fabrication, 10]   \n",
       "1    [agent based modeling, 6]  [building information modelling, 3]   \n",
       "2                         None                                 None   \n",
       "3                         None                                 None   \n",
       "4                         None                                 None   \n",
       "5                         None                                 None   \n",
       "6                         None                                 None   \n",
       "7                         None                                 None   \n",
       "8                         None                                 None   \n",
       "9                         None                                 None   \n",
       "10                        None                                 None   \n",
       "\n",
       "                                    6                             7   \\\n",
       "0   [building information modeling, 5]  [finite element analysis, 9]   \n",
       "1         [design process planning, 2]   [algorithm automatic 2d, 2]   \n",
       "2                                 None                          None   \n",
       "3                                 None                          None   \n",
       "4                                 None                          None   \n",
       "5                                 None                          None   \n",
       "6                                 None                          None   \n",
       "7                                 None                          None   \n",
       "8                                 None                          None   \n",
       "9                                 None                          None   \n",
       "10                                None                          None   \n",
       "\n",
       "                                    8   \\\n",
       "0   [geometric constraint solving, 14]   \n",
       "1    [geometric constraint systems, 6]   \n",
       "2                                 None   \n",
       "3                                 None   \n",
       "4                                 None   \n",
       "5                                 None   \n",
       "6                                 None   \n",
       "7                                 None   \n",
       "8                                 None   \n",
       "9                                 None   \n",
       "10                                None   \n",
       "\n",
       "                                             9   \\\n",
       "0          [building information modeling, 106]   \n",
       "1                [information modeling bim, 76]   \n",
       "2             [industry foundation classes, 30]   \n",
       "3   [architecture engineering construction, 24]   \n",
       "4             [building information models, 22]   \n",
       "5                  [foundation classes ifc, 21]   \n",
       "6               [semantic web technologies, 14]   \n",
       "7              [building information model, 11]   \n",
       "8                 [bv building information, 11]   \n",
       "9            [engineering construction aec, 11]   \n",
       "10          [building information modelling, 5]   \n",
       "\n",
       "                               10                                  11  \\\n",
       "0   [finite element analysis, 10]  [architectural design process, 11]   \n",
       "1         [free form surfaces, 7]    [architectural design studio, 2]   \n",
       "2                            None                                None   \n",
       "3                            None                                None   \n",
       "4                            None                                None   \n",
       "5                            None                                None   \n",
       "6                            None                                None   \n",
       "7                            None                                None   \n",
       "8                            None                                None   \n",
       "9                            None                                None   \n",
       "10                           None                                None   \n",
       "\n",
       "                                        12                              13  \\\n",
       "0      [building information modeling, 64]  [human behavior simulation, 6]   \n",
       "1           [information modeling bim, 47]   [classicism new synthesis, 2]   \n",
       "2               [tool path generation, 39]                            None   \n",
       "3             [support vector machine, 29]                            None   \n",
       "4              [computer aided design, 18]                            None   \n",
       "5               [genetic algorithm ga, 17]                            None   \n",
       "6        [particle swarm optimization, 16]                            None   \n",
       "7            [finite element analysis, 15]                            None   \n",
       "8   [experimental results demonstrate, 14]                            None   \n",
       "9              [finite element method, 14]                            None   \n",
       "10       [sculptured surface machining, 9]                            None   \n",
       "\n",
       "                                    14                              15  \\\n",
       "0   [building information modeling, 8]  [analog digital notational, 2]   \n",
       "1               [land use planning, 8]             [art net london, 2]   \n",
       "2                                 None                            None   \n",
       "3                                 None                            None   \n",
       "4                                 None                            None   \n",
       "5                                 None                            None   \n",
       "6                                 None                            None   \n",
       "7                                 None                            None   \n",
       "8                                 None                            None   \n",
       "9                                 None                            None   \n",
       "10                                None                            None   \n",
       "\n",
       "                                 16  \n",
       "0   [creative conceptual design, 9]  \n",
       "1            [case based design, 8]  \n",
       "2                              None  \n",
       "3                              None  \n",
       "4                              None  \n",
       "5                              None  \n",
       "6                              None  \n",
       "7                              None  \n",
       "8                              None  \n",
       "9                              None  \n",
       "10                             None  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(\"TOP 5 TRIGRAMS ABSTRACTS\")\n",
    "trigramScoreAbstracts = rankWords(clusterTextBagsAbstracts,phrase=3)\n",
    "display(pd.DataFrame(trigramScoreAbstracts).transpose().head())\n",
    "\n",
    "print(\"TOP 5 TRIGRAMS TITLES\")\n",
    "trigramScoreTitles = rankWords(clusterTextBagsTitles,phrase=3,longphrasescores=[[trigramScoreAbstracts,True]])\n",
    "display(pd.DataFrame(trigramScoreTitles).transpose().head())\n",
    "\n",
    "\n",
    "print(\"TOP 5 TRIGRAMS COMBINED\")\n",
    "trigrams = addClusterScores(trigramScoreAbstracts,trigramScoreTitles)\n",
    "display(pd.DataFrame(trigrams).transpose().head())\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TOP 5 BIGRAMS ABSTRACTS\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>10</th>\n",
       "      <th>11</th>\n",
       "      <th>12</th>\n",
       "      <th>13</th>\n",
       "      <th>14</th>\n",
       "      <th>15</th>\n",
       "      <th>16</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>[shape grammars, 119]</td>\n",
       "      <td>[construction sites, 55]</td>\n",
       "      <td>[space syntax, 65]</td>\n",
       "      <td>[spatial planning, 18]</td>\n",
       "      <td>[urban growth, 53]</td>\n",
       "      <td>[design process, 79]</td>\n",
       "      <td>[bim adoption, 13]</td>\n",
       "      <td>[solid model, 29]</td>\n",
       "      <td>[geometric constraint, 32]</td>\n",
       "      <td>[building information, 151]</td>\n",
       "      <td>[control points, 88]</td>\n",
       "      <td>[design process, 143]</td>\n",
       "      <td>[case study, 186]</td>\n",
       "      <td>[parametric design, 24]</td>\n",
       "      <td>[case study, 37]</td>\n",
       "      <td>[architectural design, 8]</td>\n",
       "      <td>[design process, 85]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>[shape grammar, 110]</td>\n",
       "      <td>[construction site, 54]</td>\n",
       "      <td>[street network, 21]</td>\n",
       "      <td>[planning process, 13]</td>\n",
       "      <td>[cellular automata, 29]</td>\n",
       "      <td>[digital fabrication, 70]</td>\n",
       "      <td>[case study, 12]</td>\n",
       "      <td>[mesh generation, 27]</td>\n",
       "      <td>[point cloud, 29]</td>\n",
       "      <td>[construction industry, 44]</td>\n",
       "      <td>[bspline curve, 54]</td>\n",
       "      <td>[design processes, 49]</td>\n",
       "      <td>[proposed method, 182]</td>\n",
       "      <td>[architectural design, 17]</td>\n",
       "      <td>[design process, 28]</td>\n",
       "      <td>None</td>\n",
       "      <td>[conceptual design, 68]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>[design process, 34]</td>\n",
       "      <td>[point clouds, 42]</td>\n",
       "      <td>[spatial structure, 20]</td>\n",
       "      <td>None</td>\n",
       "      <td>[landuse change, 28]</td>\n",
       "      <td>[computational design, 60]</td>\n",
       "      <td>None</td>\n",
       "      <td>[case study, 22]</td>\n",
       "      <td>[boolean operations, 27]</td>\n",
       "      <td>[case study, 42]</td>\n",
       "      <td>[data points, 48]</td>\n",
       "      <td>[architectural design, 43]</td>\n",
       "      <td>[design process, 111]</td>\n",
       "      <td>[parametric modeling, 17]</td>\n",
       "      <td>[virtual reality, 23]</td>\n",
       "      <td>None</td>\n",
       "      <td>[engineering design, 45]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>[architectural design, 33]</td>\n",
       "      <td>[construction industry, 35]</td>\n",
       "      <td>[urban design, 19]</td>\n",
       "      <td>None</td>\n",
       "      <td>[urban planning, 26]</td>\n",
       "      <td>[architectural design, 58]</td>\n",
       "      <td>None</td>\n",
       "      <td>[product configuration, 21]</td>\n",
       "      <td>[data structure, 23]</td>\n",
       "      <td>[information modeling, 32]</td>\n",
       "      <td>[curves surfaces, 45]</td>\n",
       "      <td>[design research, 33]</td>\n",
       "      <td>[tool path, 103]</td>\n",
       "      <td>[design process, 14]</td>\n",
       "      <td>[case studies, 22]</td>\n",
       "      <td>None</td>\n",
       "      <td>[case study, 29]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>[design space, 20]</td>\n",
       "      <td>[point cloud, 32]</td>\n",
       "      <td>[spatial configuration, 15]</td>\n",
       "      <td>None</td>\n",
       "      <td>[urban design, 25]</td>\n",
       "      <td>[design fabrication, 46]</td>\n",
       "      <td>None</td>\n",
       "      <td>[design process, 19]</td>\n",
       "      <td>[geometric constraints, 23]</td>\n",
       "      <td>[building design, 28]</td>\n",
       "      <td>[bezier curves, 39]</td>\n",
       "      <td>[design thinking, 31]</td>\n",
       "      <td>[construction industry, 98]</td>\n",
       "      <td>[case studies, 13]</td>\n",
       "      <td>[architectural design, 20]</td>\n",
       "      <td>None</td>\n",
       "      <td>[design knowledge, 29]</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                           0                            1   \\\n",
       "0       [shape grammars, 119]     [construction sites, 55]   \n",
       "1        [shape grammar, 110]      [construction site, 54]   \n",
       "2        [design process, 34]           [point clouds, 42]   \n",
       "3  [architectural design, 33]  [construction industry, 35]   \n",
       "4          [design space, 20]            [point cloud, 32]   \n",
       "\n",
       "                            2                       3   \\\n",
       "0           [space syntax, 65]  [spatial planning, 18]   \n",
       "1         [street network, 21]  [planning process, 13]   \n",
       "2      [spatial structure, 20]                    None   \n",
       "3           [urban design, 19]                    None   \n",
       "4  [spatial configuration, 15]                    None   \n",
       "\n",
       "                        4                           5                   6   \\\n",
       "0       [urban growth, 53]        [design process, 79]  [bim adoption, 13]   \n",
       "1  [cellular automata, 29]   [digital fabrication, 70]    [case study, 12]   \n",
       "2     [landuse change, 28]  [computational design, 60]                None   \n",
       "3     [urban planning, 26]  [architectural design, 58]                None   \n",
       "4       [urban design, 25]    [design fabrication, 46]                None   \n",
       "\n",
       "                            7                            8   \\\n",
       "0            [solid model, 29]   [geometric constraint, 32]   \n",
       "1        [mesh generation, 27]            [point cloud, 29]   \n",
       "2             [case study, 22]     [boolean operations, 27]   \n",
       "3  [product configuration, 21]         [data structure, 23]   \n",
       "4         [design process, 19]  [geometric constraints, 23]   \n",
       "\n",
       "                            9                      10  \\\n",
       "0  [building information, 151]   [control points, 88]   \n",
       "1  [construction industry, 44]    [bspline curve, 54]   \n",
       "2             [case study, 42]      [data points, 48]   \n",
       "3   [information modeling, 32]  [curves surfaces, 45]   \n",
       "4        [building design, 28]    [bezier curves, 39]   \n",
       "\n",
       "                           11                           12  \\\n",
       "0       [design process, 143]            [case study, 186]   \n",
       "1      [design processes, 49]       [proposed method, 182]   \n",
       "2  [architectural design, 43]        [design process, 111]   \n",
       "3       [design research, 33]             [tool path, 103]   \n",
       "4       [design thinking, 31]  [construction industry, 98]   \n",
       "\n",
       "                           13                          14  \\\n",
       "0     [parametric design, 24]            [case study, 37]   \n",
       "1  [architectural design, 17]        [design process, 28]   \n",
       "2   [parametric modeling, 17]       [virtual reality, 23]   \n",
       "3        [design process, 14]          [case studies, 22]   \n",
       "4          [case studies, 13]  [architectural design, 20]   \n",
       "\n",
       "                          15                        16  \n",
       "0  [architectural design, 8]      [design process, 85]  \n",
       "1                       None   [conceptual design, 68]  \n",
       "2                       None  [engineering design, 45]  \n",
       "3                       None          [case study, 29]  \n",
       "4                       None    [design knowledge, 29]  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TOP 5 BIGRAMS TITLES\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>10</th>\n",
       "      <th>11</th>\n",
       "      <th>12</th>\n",
       "      <th>13</th>\n",
       "      <th>14</th>\n",
       "      <th>15</th>\n",
       "      <th>16</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>[generative design, 11]</td>\n",
       "      <td>[real time, 15]</td>\n",
       "      <td>[analysis urban, 7]</td>\n",
       "      <td>[planning design, 6]</td>\n",
       "      <td>[planning support, 17]</td>\n",
       "      <td>[robotic fabrication, 12]</td>\n",
       "      <td>[product data, 4]</td>\n",
       "      <td>[design rationale, 10]</td>\n",
       "      <td>[surface reconstruction, 14]</td>\n",
       "      <td>[bim based, 19]</td>\n",
       "      <td>[spline curves, 21]</td>\n",
       "      <td>[conceptual design, 10]</td>\n",
       "      <td>[real time, 28]</td>\n",
       "      <td>[design space, 5]</td>\n",
       "      <td>[computer aided, 8]</td>\n",
       "      <td>[modern architecture, 3]</td>\n",
       "      <td>[case based, 13]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[form finding, 11]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[ontology based, 13]</td>\n",
       "      <td>[spline surfaces, 19]</td>\n",
       "      <td>None</td>\n",
       "      <td>[construction projects, 18]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[nurbs surfaces, 14]</td>\n",
       "      <td>None</td>\n",
       "      <td>[path planning, 17]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[subdivision surfaces, 14]</td>\n",
       "      <td>None</td>\n",
       "      <td>[augmented reality, 16]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[spline surface, 13]</td>\n",
       "      <td>None</td>\n",
       "      <td>[based approach, 16]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                        0                1                    2   \\\n",
       "0  [generative design, 11]  [real time, 15]  [analysis urban, 7]   \n",
       "1                     None             None                 None   \n",
       "2                     None             None                 None   \n",
       "3                     None             None                 None   \n",
       "4                     None             None                 None   \n",
       "\n",
       "                     3                       4                          5   \\\n",
       "0  [planning design, 6]  [planning support, 17]  [robotic fabrication, 12]   \n",
       "1                  None                    None         [form finding, 11]   \n",
       "2                  None                    None                       None   \n",
       "3                  None                    None                       None   \n",
       "4                  None                    None                       None   \n",
       "\n",
       "                  6                       7                             8   \\\n",
       "0  [product data, 4]  [design rationale, 10]  [surface reconstruction, 14]   \n",
       "1               None                    None                          None   \n",
       "2               None                    None                          None   \n",
       "3               None                    None                          None   \n",
       "4               None                    None                          None   \n",
       "\n",
       "                     9                           10                       11  \\\n",
       "0       [bim based, 19]         [spline curves, 21]  [conceptual design, 10]   \n",
       "1  [ontology based, 13]       [spline surfaces, 19]                     None   \n",
       "2                  None        [nurbs surfaces, 14]                     None   \n",
       "3                  None  [subdivision surfaces, 14]                     None   \n",
       "4                  None        [spline surface, 13]                     None   \n",
       "\n",
       "                            12                 13                   14  \\\n",
       "0              [real time, 28]  [design space, 5]  [computer aided, 8]   \n",
       "1  [construction projects, 18]               None                 None   \n",
       "2          [path planning, 17]               None                 None   \n",
       "3      [augmented reality, 16]               None                 None   \n",
       "4         [based approach, 16]               None                 None   \n",
       "\n",
       "                         15                16  \n",
       "0  [modern architecture, 3]  [case based, 13]  \n",
       "1                      None              None  \n",
       "2                      None              None  \n",
       "3                      None              None  \n",
       "4                      None              None  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TOP 5 BIGRAMS COMBINED\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>10</th>\n",
       "      <th>11</th>\n",
       "      <th>12</th>\n",
       "      <th>13</th>\n",
       "      <th>14</th>\n",
       "      <th>15</th>\n",
       "      <th>16</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>[shape grammars, 119]</td>\n",
       "      <td>[construction sites, 55]</td>\n",
       "      <td>[space syntax, 65]</td>\n",
       "      <td>[spatial planning, 18]</td>\n",
       "      <td>[urban growth, 53]</td>\n",
       "      <td>[design process, 79]</td>\n",
       "      <td>[bim adoption, 13]</td>\n",
       "      <td>[solid model, 29]</td>\n",
       "      <td>[geometric constraint, 32]</td>\n",
       "      <td>[building information, 151]</td>\n",
       "      <td>[control points, 88]</td>\n",
       "      <td>[design process, 143]</td>\n",
       "      <td>[case study, 186]</td>\n",
       "      <td>[parametric design, 24]</td>\n",
       "      <td>[case study, 37]</td>\n",
       "      <td>[architectural design, 8]</td>\n",
       "      <td>[design process, 85]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>[shape grammar, 110]</td>\n",
       "      <td>[construction site, 54]</td>\n",
       "      <td>[street network, 21]</td>\n",
       "      <td>[planning process, 13]</td>\n",
       "      <td>[cellular automata, 29]</td>\n",
       "      <td>[digital fabrication, 70]</td>\n",
       "      <td>[case study, 12]</td>\n",
       "      <td>[mesh generation, 27]</td>\n",
       "      <td>[point cloud, 29]</td>\n",
       "      <td>[construction industry, 44]</td>\n",
       "      <td>[bspline curve, 54]</td>\n",
       "      <td>[design processes, 49]</td>\n",
       "      <td>[proposed method, 182]</td>\n",
       "      <td>[architectural design, 17]</td>\n",
       "      <td>[design process, 28]</td>\n",
       "      <td>[modern architecture, 3]</td>\n",
       "      <td>[conceptual design, 68]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>[design process, 34]</td>\n",
       "      <td>[point clouds, 42]</td>\n",
       "      <td>[spatial structure, 20]</td>\n",
       "      <td>[planning design, 6]</td>\n",
       "      <td>[landuse change, 28]</td>\n",
       "      <td>[computational design, 60]</td>\n",
       "      <td>[product data, 4]</td>\n",
       "      <td>[case study, 22]</td>\n",
       "      <td>[boolean operations, 27]</td>\n",
       "      <td>[case study, 42]</td>\n",
       "      <td>[data points, 48]</td>\n",
       "      <td>[architectural design, 43]</td>\n",
       "      <td>[design process, 111]</td>\n",
       "      <td>[parametric modeling, 17]</td>\n",
       "      <td>[virtual reality, 23]</td>\n",
       "      <td>None</td>\n",
       "      <td>[engineering design, 45]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>[architectural design, 33]</td>\n",
       "      <td>[construction industry, 35]</td>\n",
       "      <td>[urban design, 19]</td>\n",
       "      <td>None</td>\n",
       "      <td>[urban planning, 26]</td>\n",
       "      <td>[architectural design, 58]</td>\n",
       "      <td>None</td>\n",
       "      <td>[product configuration, 21]</td>\n",
       "      <td>[data structure, 23]</td>\n",
       "      <td>[information modeling, 32]</td>\n",
       "      <td>[curves surfaces, 45]</td>\n",
       "      <td>[design research, 33]</td>\n",
       "      <td>[tool path, 103]</td>\n",
       "      <td>[design process, 14]</td>\n",
       "      <td>[case studies, 22]</td>\n",
       "      <td>None</td>\n",
       "      <td>[case study, 29]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>[design space, 20]</td>\n",
       "      <td>[point cloud, 32]</td>\n",
       "      <td>[spatial configuration, 15]</td>\n",
       "      <td>None</td>\n",
       "      <td>[urban design, 25]</td>\n",
       "      <td>[design fabrication, 46]</td>\n",
       "      <td>None</td>\n",
       "      <td>[design process, 19]</td>\n",
       "      <td>[geometric constraints, 23]</td>\n",
       "      <td>[building design, 28]</td>\n",
       "      <td>[bezier curves, 39]</td>\n",
       "      <td>[design thinking, 31]</td>\n",
       "      <td>[construction industry, 98]</td>\n",
       "      <td>[case studies, 13]</td>\n",
       "      <td>[architectural design, 20]</td>\n",
       "      <td>None</td>\n",
       "      <td>[design knowledge, 29]</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                           0                            1   \\\n",
       "0       [shape grammars, 119]     [construction sites, 55]   \n",
       "1        [shape grammar, 110]      [construction site, 54]   \n",
       "2        [design process, 34]           [point clouds, 42]   \n",
       "3  [architectural design, 33]  [construction industry, 35]   \n",
       "4          [design space, 20]            [point cloud, 32]   \n",
       "\n",
       "                            2                       3   \\\n",
       "0           [space syntax, 65]  [spatial planning, 18]   \n",
       "1         [street network, 21]  [planning process, 13]   \n",
       "2      [spatial structure, 20]    [planning design, 6]   \n",
       "3           [urban design, 19]                    None   \n",
       "4  [spatial configuration, 15]                    None   \n",
       "\n",
       "                        4                           5                   6   \\\n",
       "0       [urban growth, 53]        [design process, 79]  [bim adoption, 13]   \n",
       "1  [cellular automata, 29]   [digital fabrication, 70]    [case study, 12]   \n",
       "2     [landuse change, 28]  [computational design, 60]   [product data, 4]   \n",
       "3     [urban planning, 26]  [architectural design, 58]                None   \n",
       "4       [urban design, 25]    [design fabrication, 46]                None   \n",
       "\n",
       "                            7                            8   \\\n",
       "0            [solid model, 29]   [geometric constraint, 32]   \n",
       "1        [mesh generation, 27]            [point cloud, 29]   \n",
       "2             [case study, 22]     [boolean operations, 27]   \n",
       "3  [product configuration, 21]         [data structure, 23]   \n",
       "4         [design process, 19]  [geometric constraints, 23]   \n",
       "\n",
       "                            9                      10  \\\n",
       "0  [building information, 151]   [control points, 88]   \n",
       "1  [construction industry, 44]    [bspline curve, 54]   \n",
       "2             [case study, 42]      [data points, 48]   \n",
       "3   [information modeling, 32]  [curves surfaces, 45]   \n",
       "4        [building design, 28]    [bezier curves, 39]   \n",
       "\n",
       "                           11                           12  \\\n",
       "0       [design process, 143]            [case study, 186]   \n",
       "1      [design processes, 49]       [proposed method, 182]   \n",
       "2  [architectural design, 43]        [design process, 111]   \n",
       "3       [design research, 33]             [tool path, 103]   \n",
       "4       [design thinking, 31]  [construction industry, 98]   \n",
       "\n",
       "                           13                          14  \\\n",
       "0     [parametric design, 24]            [case study, 37]   \n",
       "1  [architectural design, 17]        [design process, 28]   \n",
       "2   [parametric modeling, 17]       [virtual reality, 23]   \n",
       "3        [design process, 14]          [case studies, 22]   \n",
       "4          [case studies, 13]  [architectural design, 20]   \n",
       "\n",
       "                          15                        16  \n",
       "0  [architectural design, 8]      [design process, 85]  \n",
       "1   [modern architecture, 3]   [conceptual design, 68]  \n",
       "2                       None  [engineering design, 45]  \n",
       "3                       None          [case study, 29]  \n",
       "4                       None    [design knowledge, 29]  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(\"TOP 5 BIGRAMS ABSTRACTS\")\n",
    "bigramScoreAbstracts = rankWords(clusterTextBagsAbstracts,phrase=2, n=5,longphrasescores=[[trigrams,False]])\n",
    "#bigramScoreAbstracts = accountScoreClusters(trigramScoreAbstracts, bigramScoreAbstracts)\n",
    "display(pd.DataFrame(bigramScoreAbstracts).transpose().head())\n",
    "\n",
    "print(\"TOP 5 BIGRAMS TITLES\")\n",
    "bigramScoreTitles = rankWords(clusterTextBagsTitles,phrase=2, n=5,longphrasescores=[[trigrams,False],[bigramScoreAbstracts,True]])\n",
    "#print(bigramScoreTitles[:10])\n",
    "#bigramScoreTitles = accountScoreClusters(trigramScoreTitles, bigramScoreTitles)\n",
    "#print(bigramScoreTitles[:10])\n",
    "display(pd.DataFrame(bigramScoreTitles).transpose().head())\n",
    "\n",
    "print(\"TOP 5 BIGRAMS COMBINED\")\n",
    "bigrams = addClusterScores(bigramScoreAbstracts,bigramScoreTitles)\n",
    "display(pd.DataFrame(bigrams).transpose().head())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 131,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ABSTRACTS\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>10</th>\n",
       "      <th>11</th>\n",
       "      <th>12</th>\n",
       "      <th>13</th>\n",
       "      <th>14</th>\n",
       "      <th>15</th>\n",
       "      <th>16</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>[design, 655]</td>\n",
       "      <td>[construction, 491]</td>\n",
       "      <td>[urban, 208]</td>\n",
       "      <td>[planning, 281]</td>\n",
       "      <td>[urban, 437]</td>\n",
       "      <td>[design, 885]</td>\n",
       "      <td>[design, 167]</td>\n",
       "      <td>[design, 349]</td>\n",
       "      <td>[geometric, 236]</td>\n",
       "      <td>[bim, 419]</td>\n",
       "      <td>[surface, 596]</td>\n",
       "      <td>[design, 1539]</td>\n",
       "      <td>[design, 1520]</td>\n",
       "      <td>[design, 310]</td>\n",
       "      <td>[design, 521]</td>\n",
       "      <td>[design, 109]</td>\n",
       "      <td>[design, 945]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>[shape, 271]</td>\n",
       "      <td>[data, 227]</td>\n",
       "      <td>[spatial, 196]</td>\n",
       "      <td>[design, 137]</td>\n",
       "      <td>[model, 388]</td>\n",
       "      <td>[fabrication, 285]</td>\n",
       "      <td>[model, 97]</td>\n",
       "      <td>[model, 191]</td>\n",
       "      <td>[method, 228]</td>\n",
       "      <td>[information, 402]</td>\n",
       "      <td>[surfaces, 454]</td>\n",
       "      <td>[designers, 310]</td>\n",
       "      <td>[construction, 1319]</td>\n",
       "      <td>[model, 65]</td>\n",
       "      <td>[model, 260]</td>\n",
       "      <td>[urban, 44]</td>\n",
       "      <td>[function, 198]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>[grammar, 158]</td>\n",
       "      <td>[information, 166]</td>\n",
       "      <td>[analysis, 163]</td>\n",
       "      <td>[development, 67]</td>\n",
       "      <td>[planning, 214]</td>\n",
       "      <td>[research, 272]</td>\n",
       "      <td>[construction, 84]</td>\n",
       "      <td>[mesh, 178]</td>\n",
       "      <td>[model, 202]</td>\n",
       "      <td>[building, 374]</td>\n",
       "      <td>[curves, 431]</td>\n",
       "      <td>[study, 228]</td>\n",
       "      <td>[model, 1308]</td>\n",
       "      <td>[data, 58]</td>\n",
       "      <td>[data, 217]</td>\n",
       "      <td>[architecture, 42]</td>\n",
       "      <td>[knowledge, 161]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>[rules, 140]</td>\n",
       "      <td>[method, 159]</td>\n",
       "      <td>[space, 135]</td>\n",
       "      <td>[urban, 63]</td>\n",
       "      <td>[spatial, 189]</td>\n",
       "      <td>[material, 249]</td>\n",
       "      <td>[data, 79]</td>\n",
       "      <td>[method, 172]</td>\n",
       "      <td>[surface, 180]</td>\n",
       "      <td>[design, 364]</td>\n",
       "      <td>[method, 387]</td>\n",
       "      <td>[research, 156]</td>\n",
       "      <td>[data, 1084]</td>\n",
       "      <td>[parametric, 56]</td>\n",
       "      <td>[process, 187]</td>\n",
       "      <td>[research, 39]</td>\n",
       "      <td>[model, 158]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>[shapes, 113]</td>\n",
       "      <td>[safety, 156]</td>\n",
       "      <td>[design, 111]</td>\n",
       "      <td>[local, 60]</td>\n",
       "      <td>[models, 175]</td>\n",
       "      <td>[digital, 219]</td>\n",
       "      <td>[approach, 66]</td>\n",
       "      <td>[approach, 171]</td>\n",
       "      <td>[algorithm, 175]</td>\n",
       "      <td>[construction, 308]</td>\n",
       "      <td>[curve, 344]</td>\n",
       "      <td>[process, 143]</td>\n",
       "      <td>[method, 1064]</td>\n",
       "      <td>[approach, 52]</td>\n",
       "      <td>[approach, 186]</td>\n",
       "      <td>[process, 31]</td>\n",
       "      <td>[approach, 146]</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               0                    1                2                  3   \\\n",
       "0   [design, 655]  [construction, 491]     [urban, 208]    [planning, 281]   \n",
       "1    [shape, 271]          [data, 227]   [spatial, 196]      [design, 137]   \n",
       "2  [grammar, 158]   [information, 166]  [analysis, 163]  [development, 67]   \n",
       "3    [rules, 140]        [method, 159]     [space, 135]        [urban, 63]   \n",
       "4   [shapes, 113]        [safety, 156]    [design, 111]        [local, 60]   \n",
       "\n",
       "                4                   5                   6                7   \\\n",
       "0     [urban, 437]       [design, 885]       [design, 167]    [design, 349]   \n",
       "1     [model, 388]  [fabrication, 285]         [model, 97]     [model, 191]   \n",
       "2  [planning, 214]     [research, 272]  [construction, 84]      [mesh, 178]   \n",
       "3   [spatial, 189]     [material, 249]          [data, 79]    [method, 172]   \n",
       "4    [models, 175]      [digital, 219]      [approach, 66]  [approach, 171]   \n",
       "\n",
       "                 8                    9                10                11  \\\n",
       "0  [geometric, 236]           [bim, 419]   [surface, 596]    [design, 1539]   \n",
       "1     [method, 228]   [information, 402]  [surfaces, 454]  [designers, 310]   \n",
       "2      [model, 202]      [building, 374]    [curves, 431]      [study, 228]   \n",
       "3    [surface, 180]        [design, 364]    [method, 387]   [research, 156]   \n",
       "4  [algorithm, 175]  [construction, 308]     [curve, 344]    [process, 143]   \n",
       "\n",
       "                     12                13               14  \\\n",
       "0        [design, 1520]     [design, 310]    [design, 521]   \n",
       "1  [construction, 1319]       [model, 65]     [model, 260]   \n",
       "2         [model, 1308]        [data, 58]      [data, 217]   \n",
       "3          [data, 1084]  [parametric, 56]   [process, 187]   \n",
       "4        [method, 1064]    [approach, 52]  [approach, 186]   \n",
       "\n",
       "                   15                16  \n",
       "0       [design, 109]     [design, 945]  \n",
       "1         [urban, 44]   [function, 198]  \n",
       "2  [architecture, 42]  [knowledge, 161]  \n",
       "3      [research, 39]      [model, 158]  \n",
       "4       [process, 31]   [approach, 146]  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TITLES\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>10</th>\n",
       "      <th>11</th>\n",
       "      <th>12</th>\n",
       "      <th>13</th>\n",
       "      <th>14</th>\n",
       "      <th>15</th>\n",
       "      <th>16</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>[architecture, 21]</td>\n",
       "      <td>[3d, 33]</td>\n",
       "      <td>[approach, 12]</td>\n",
       "      <td>[case, 12]</td>\n",
       "      <td>[based, 31]</td>\n",
       "      <td>[robotic, 23]</td>\n",
       "      <td>[systems, 13]</td>\n",
       "      <td>[modeling, 20]</td>\n",
       "      <td>[analysis, 21]</td>\n",
       "      <td>[based, 52]</td>\n",
       "      <td>[spline, 81]</td>\n",
       "      <td>[engineering, 21]</td>\n",
       "      <td>[simulation, 96]</td>\n",
       "      <td>[planning, 17]</td>\n",
       "      <td>[architecture, 28]</td>\n",
       "      <td>[new, 10]</td>\n",
       "      <td>[based, 41]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>[language, 17]</td>\n",
       "      <td>[automated, 30]</td>\n",
       "      <td>[morphology, 11]</td>\n",
       "      <td>None</td>\n",
       "      <td>[simulation, 22]</td>\n",
       "      <td>[systems, 21]</td>\n",
       "      <td>[simulation, 11]</td>\n",
       "      <td>[building, 14]</td>\n",
       "      <td>[cad, 20]</td>\n",
       "      <td>[analysis, 19]</td>\n",
       "      <td>[interpolation, 45]</td>\n",
       "      <td>[designing, 19]</td>\n",
       "      <td>[modeling, 86]</td>\n",
       "      <td>[urban, 12]</td>\n",
       "      <td>[decision, 22]</td>\n",
       "      <td>None</td>\n",
       "      <td>[learning, 13]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>[plans, 15]</td>\n",
       "      <td>[laser, 15]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[cities, 18]</td>\n",
       "      <td>[performance, 19]</td>\n",
       "      <td>None</td>\n",
       "      <td>[meshing, 14]</td>\n",
       "      <td>[3d, 19]</td>\n",
       "      <td>[architectural, 18]</td>\n",
       "      <td>[splines, 43]</td>\n",
       "      <td>[creative, 18]</td>\n",
       "      <td>[3d, 84]</td>\n",
       "      <td>None</td>\n",
       "      <td>[modeling, 21]</td>\n",
       "      <td>None</td>\n",
       "      <td>[planning, 13]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>[implementation, 14]</td>\n",
       "      <td>[monitoring, 15]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[gis, 17]</td>\n",
       "      <td>[parametric, 18]</td>\n",
       "      <td>None</td>\n",
       "      <td>[3d, 12]</td>\n",
       "      <td>[solid, 19]</td>\n",
       "      <td>[automated, 16]</td>\n",
       "      <td>[approximation, 34]</td>\n",
       "      <td>[digital, 18]</td>\n",
       "      <td>[management, 77]</td>\n",
       "      <td>None</td>\n",
       "      <td>[systems, 21]</td>\n",
       "      <td>None</td>\n",
       "      <td>[reasoning, 13]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>[planning, 14]</td>\n",
       "      <td>[bim, 14]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[dynamics, 15]</td>\n",
       "      <td>[concrete, 15]</td>\n",
       "      <td>None</td>\n",
       "      <td>[engineering, 12]</td>\n",
       "      <td>[surfaces, 19]</td>\n",
       "      <td>[performance, 16]</td>\n",
       "      <td>[modeling, 34]</td>\n",
       "      <td>[systems, 17]</td>\n",
       "      <td>[machining, 75]</td>\n",
       "      <td>None</td>\n",
       "      <td>[simulation, 20]</td>\n",
       "      <td>None</td>\n",
       "      <td>[urban, 13]</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     0                 1                 2           3   \\\n",
       "0    [architecture, 21]          [3d, 33]    [approach, 12]  [case, 12]   \n",
       "1        [language, 17]   [automated, 30]  [morphology, 11]        None   \n",
       "2           [plans, 15]       [laser, 15]              None        None   \n",
       "3  [implementation, 14]  [monitoring, 15]              None        None   \n",
       "4        [planning, 14]         [bim, 14]              None        None   \n",
       "\n",
       "                 4                  5                 6                  7   \\\n",
       "0       [based, 31]      [robotic, 23]     [systems, 13]     [modeling, 20]   \n",
       "1  [simulation, 22]      [systems, 21]  [simulation, 11]     [building, 14]   \n",
       "2      [cities, 18]  [performance, 19]              None      [meshing, 14]   \n",
       "3         [gis, 17]   [parametric, 18]              None           [3d, 12]   \n",
       "4    [dynamics, 15]     [concrete, 15]              None  [engineering, 12]   \n",
       "\n",
       "               8                    9                    10  \\\n",
       "0  [analysis, 21]          [based, 52]         [spline, 81]   \n",
       "1       [cad, 20]       [analysis, 19]  [interpolation, 45]   \n",
       "2        [3d, 19]  [architectural, 18]        [splines, 43]   \n",
       "3     [solid, 19]      [automated, 16]  [approximation, 34]   \n",
       "4  [surfaces, 19]    [performance, 16]       [modeling, 34]   \n",
       "\n",
       "                  11                12              13                  14  \\\n",
       "0  [engineering, 21]  [simulation, 96]  [planning, 17]  [architecture, 28]   \n",
       "1    [designing, 19]    [modeling, 86]     [urban, 12]      [decision, 22]   \n",
       "2     [creative, 18]          [3d, 84]            None      [modeling, 21]   \n",
       "3      [digital, 18]  [management, 77]            None       [systems, 21]   \n",
       "4      [systems, 17]   [machining, 75]            None    [simulation, 20]   \n",
       "\n",
       "          15               16  \n",
       "0  [new, 10]      [based, 41]  \n",
       "1       None   [learning, 13]  \n",
       "2       None   [planning, 13]  \n",
       "3       None  [reasoning, 13]  \n",
       "4       None      [urban, 13]  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "print(\"TOP 5 WORDS IN ABSTRACTS\")\n",
    "wordScoreAbstracts = rankWords(clusterTextBagsAbstracts,phrase=1, n=20, longphrasescores=[[trigrams,False],[bigrams,False]])\n",
    "display(pd.DataFrame(wordScoreAbstracts).transpose().head())\n",
    "\n",
    "print(\"TOP 5 WORDS IN TITLES\")\n",
    "wordScoreTitles = rankWords(clusterTextBagsTitles,phrase=1, n=20, longphrasescores=[[trigrams,False],[bigrams,False],[wordScoreAbstracts,True]])\n",
    "display(pd.DataFrame(wordScoreTitles).transpose().head())\n",
    "\n",
    "words = addClusterScores(wordScoreAbstracts,wordScoreTitles)\n",
    "#display(pd.DataFrame(words).transpose())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 130,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TOP 5 WORDS\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>10</th>\n",
       "      <th>11</th>\n",
       "      <th>12</th>\n",
       "      <th>13</th>\n",
       "      <th>14</th>\n",
       "      <th>15</th>\n",
       "      <th>16</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>[shape, 271]</td>\n",
       "      <td>[construction, 491]</td>\n",
       "      <td>[urban, 208]</td>\n",
       "      <td>[planning, 281]</td>\n",
       "      <td>[urban, 437]</td>\n",
       "      <td>[fabrication, 285]</td>\n",
       "      <td>[construction, 84]</td>\n",
       "      <td>[mesh, 178]</td>\n",
       "      <td>[geometric, 236]</td>\n",
       "      <td>[bim, 419]</td>\n",
       "      <td>[surface, 596]</td>\n",
       "      <td>[designers, 310]</td>\n",
       "      <td>[construction, 1319]</td>\n",
       "      <td>[parametric, 56]</td>\n",
       "      <td>[construction, 172]</td>\n",
       "      <td>[urban, 44]</td>\n",
       "      <td>[function, 198]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>[grammar, 158]</td>\n",
       "      <td>[safety, 156]</td>\n",
       "      <td>[patterns, 85]</td>\n",
       "      <td>[urban, 63]</td>\n",
       "      <td>[planning, 214]</td>\n",
       "      <td>[material, 249]</td>\n",
       "      <td>[developed, 52]</td>\n",
       "      <td>[configuration, 129]</td>\n",
       "      <td>[surface, 180]</td>\n",
       "      <td>[construction, 308]</td>\n",
       "      <td>[surfaces, 454]</td>\n",
       "      <td>[knowledge, 120]</td>\n",
       "      <td>[algorithm, 777]</td>\n",
       "      <td>[systems, 50]</td>\n",
       "      <td>[planning, 151]</td>\n",
       "      <td>[planning, 22]</td>\n",
       "      <td>[knowledge, 161]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>[rules, 140]</td>\n",
       "      <td>[equipment, 132]</td>\n",
       "      <td>[movement, 68]</td>\n",
       "      <td>[local, 60]</td>\n",
       "      <td>[land, 109]</td>\n",
       "      <td>[construction, 207]</td>\n",
       "      <td>[planning, 48]</td>\n",
       "      <td>[algorithm, 121]</td>\n",
       "      <td>[algorithm, 175]</td>\n",
       "      <td>[project, 200]</td>\n",
       "      <td>[curves, 431]</td>\n",
       "      <td>[students, 118]</td>\n",
       "      <td>[surface, 661]</td>\n",
       "      <td>[performance, 46]</td>\n",
       "      <td>[urban, 150]</td>\n",
       "      <td>[physical, 21]</td>\n",
       "      <td>[product, 119]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>[shapes, 113]</td>\n",
       "      <td>[tracking, 117]</td>\n",
       "      <td>[network, 65]</td>\n",
       "      <td>[plans, 53]</td>\n",
       "      <td>[city, 105]</td>\n",
       "      <td>[tools, 145]</td>\n",
       "      <td>[project, 43]</td>\n",
       "      <td>[knowledge, 112]</td>\n",
       "      <td>[set, 121]</td>\n",
       "      <td>[management, 173]</td>\n",
       "      <td>[curve, 344]</td>\n",
       "      <td>[cognitive, 113]</td>\n",
       "      <td>[developed, 626]</td>\n",
       "      <td>[tools, 46]</td>\n",
       "      <td>[project, 128]</td>\n",
       "      <td>[systems, 20]</td>\n",
       "      <td>[systems, 119]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>[grammars, 91]</td>\n",
       "      <td>[workers, 106]</td>\n",
       "      <td>[properties, 62]</td>\n",
       "      <td>[performance, 48]</td>\n",
       "      <td>[systems, 104]</td>\n",
       "      <td>[structures, 137]</td>\n",
       "      <td>[framework, 38]</td>\n",
       "      <td>[construction, 100]</td>\n",
       "      <td>[constraints, 115]</td>\n",
       "      <td>[framework, 107]</td>\n",
       "      <td>[points, 269]</td>\n",
       "      <td>[problem, 100]</td>\n",
       "      <td>[performance, 558]</td>\n",
       "      <td>[designers, 43]</td>\n",
       "      <td>[problem, 123]</td>\n",
       "      <td>[cities, 16]</td>\n",
       "      <td>[functional, 106]</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               0                    1                 2                  3   \\\n",
       "0    [shape, 271]  [construction, 491]      [urban, 208]    [planning, 281]   \n",
       "1  [grammar, 158]        [safety, 156]    [patterns, 85]        [urban, 63]   \n",
       "2    [rules, 140]     [equipment, 132]    [movement, 68]        [local, 60]   \n",
       "3   [shapes, 113]      [tracking, 117]     [network, 65]        [plans, 53]   \n",
       "4  [grammars, 91]       [workers, 106]  [properties, 62]  [performance, 48]   \n",
       "\n",
       "                4                    5                   6   \\\n",
       "0     [urban, 437]   [fabrication, 285]  [construction, 84]   \n",
       "1  [planning, 214]      [material, 249]     [developed, 52]   \n",
       "2      [land, 109]  [construction, 207]      [planning, 48]   \n",
       "3      [city, 105]         [tools, 145]       [project, 43]   \n",
       "4   [systems, 104]    [structures, 137]     [framework, 38]   \n",
       "\n",
       "                     7                   8                    9   \\\n",
       "0           [mesh, 178]    [geometric, 236]           [bim, 419]   \n",
       "1  [configuration, 129]      [surface, 180]  [construction, 308]   \n",
       "2      [algorithm, 121]    [algorithm, 175]       [project, 200]   \n",
       "3      [knowledge, 112]          [set, 121]    [management, 173]   \n",
       "4   [construction, 100]  [constraints, 115]     [framework, 107]   \n",
       "\n",
       "                10                11                    12                 13  \\\n",
       "0   [surface, 596]  [designers, 310]  [construction, 1319]   [parametric, 56]   \n",
       "1  [surfaces, 454]  [knowledge, 120]      [algorithm, 777]      [systems, 50]   \n",
       "2    [curves, 431]   [students, 118]        [surface, 661]  [performance, 46]   \n",
       "3     [curve, 344]  [cognitive, 113]      [developed, 626]        [tools, 46]   \n",
       "4    [points, 269]    [problem, 100]    [performance, 558]    [designers, 43]   \n",
       "\n",
       "                    14              15                 16  \n",
       "0  [construction, 172]     [urban, 44]    [function, 198]  \n",
       "1      [planning, 151]  [planning, 22]   [knowledge, 161]  \n",
       "2         [urban, 150]  [physical, 21]     [product, 119]  \n",
       "3       [project, 128]   [systems, 20]     [systems, 119]  \n",
       "4       [problem, 123]    [cities, 16]  [functional, 106]  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#common words in the field to skip only in the single words lists\n",
    "# ideally this should be automated to kick out all words that are occuring in more than 40-50% of clusters\n",
    "commonFieldTerms = [\n",
    "    'method', 'methods', 'process', 'design', 'designs',\n",
    "    'research', 'based', 'approach',\n",
    "    'information', 'architecture', 'architectural',\n",
    "    'digital', 'development', 'building',\n",
    "    'model', 'models', 'modeling',\n",
    "    'result', 'results', 'study', 'data',\n",
    "    'proposed', 'analysis', 'different',\n",
    "    'space', 'spaces', 'spatial', 'bv'\n",
    "]\n",
    "\n",
    "def filterScoreList(scorelist, commonTerms):\n",
    "    newscorelist = []\n",
    "    for wscore in scorelist:\n",
    "        #print(wscore)\n",
    "        if not any(wscore[0] in s for s in commonTerms):\n",
    "            newscorelist.append(wscore)\n",
    "    return newscorelist\n",
    "\n",
    "\n",
    "filteredwords = []\n",
    "for wordscores in words:\n",
    "    filteredwords.append(filterScoreList(wordscores, commonFieldTerms))\n",
    "print(\"TOP 5 WORDS\")\n",
    "display(pd.DataFrame(filteredwords).transpose().head())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now let's combine all words and phrases into the cluster labels."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 281,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[[['shape grammar interpreter', 10],\n",
       "  ['early stages design', 3],\n",
       "  ['shape grammars', 119],\n",
       "  ['shape grammar', 110],\n",
       "  ['design process', 34],\n",
       "  ['architectural design', 33],\n",
       "  ['design space', 20],\n",
       "  ['generative design', 11],\n",
       "  ['shape', 271],\n",
       "  ['grammar', 158],\n",
       "  ['rules', 140],\n",
       "  ['shapes', 113],\n",
       "  ['grammars', 91],\n",
       "  ['generation', 74],\n",
       "  ['presented', 73],\n",
       "  ['systems', 70],\n",
       "  ['language', 17],\n",
       "  ['plans', 15],\n",
       "  ['implementation', 14],\n",
       "  ['planning', 14],\n",
       "  ['formal', 13],\n",
       "  ['layout', 13],\n",
       "  ['representation', 13],\n",
       "  ['computer', 12],\n",
       "  ['floor', 12],\n",
       "  ['urban', 12],\n",
       "  ['generative', 11]],\n",
       " [['point cloud data', 17],\n",
       "  ['radio frequency identification', 16],\n",
       "  ['building information models', 13],\n",
       "  ['precast concrete elements', 12],\n",
       "  ['building information modeling', 11],\n",
       "  ['3d laser scanning', 4],\n",
       "  ['construction sites', 55],\n",
       "  ['construction site', 54],\n",
       "  ['point clouds', 42],\n",
       "  ['construction industry', 35],\n",
       "  ['point cloud', 32],\n",
       "  ['real time', 15],\n",
       "  ['construction', 491],\n",
       "  ['safety', 156],\n",
       "  ['equipment', 132],\n",
       "  ['tracking', 117],\n",
       "  ['workers', 106],\n",
       "  ['management', 98],\n",
       "  ['project', 93],\n",
       "  ['technology', 87],\n",
       "  ['3d', 33],\n",
       "  ['automated', 30],\n",
       "  ['laser', 15],\n",
       "  ['monitoring', 15],\n",
       "  ['bim', 14],\n",
       "  ['control', 14],\n",
       "  ['automatic', 13],\n",
       "  ['rfid', 13],\n",
       "  ['vision', 12],\n",
       "  ['assessment', 11],\n",
       "  ['detection', 11],\n",
       "  ['performance', 11]],\n",
       " [['space syntax analysis', 8],\n",
       "  ['urban street networks', 3],\n",
       "  ['space syntax', 65],\n",
       "  ['street network', 21],\n",
       "  ['spatial structure', 20],\n",
       "  ['urban design', 19],\n",
       "  ['spatial configuration', 15],\n",
       "  ['analysis urban', 7],\n",
       "  ['urban', 208],\n",
       "  ['patterns', 85],\n",
       "  ['movement', 68],\n",
       "  ['network', 65],\n",
       "  ['properties', 62],\n",
       "  ['measures', 57],\n",
       "  ['networks', 55],\n",
       "  ['morphology', 11]],\n",
       " [['planning support systems', 10],\n",
       "  ['positive theory planning', 3],\n",
       "  ['spatial planning', 18],\n",
       "  ['planning process', 13],\n",
       "  ['planning design', 6],\n",
       "  ['planning', 281],\n",
       "  ['urban', 63],\n",
       "  ['local', 60],\n",
       "  ['plans', 53],\n",
       "  ['performance', 48],\n",
       "  ['public', 46],\n",
       "  ['practice', 43],\n",
       "  ['plan', 42],\n",
       "  ['implementation', 40],\n",
       "  ['planners', 33],\n",
       "  ['case', 12]],\n",
       " [['cellular automata ca', 17],\n",
       "  ['agent based modeling', 6],\n",
       "  ['urban growth', 53],\n",
       "  ['cellular automata', 29],\n",
       "  ['landuse change', 28],\n",
       "  ['urban planning', 26],\n",
       "  ['urban design', 25],\n",
       "  ['planning support', 17],\n",
       "  ['urban', 437],\n",
       "  ['planning', 214],\n",
       "  ['land', 109],\n",
       "  ['city', 105],\n",
       "  ['systems', 104],\n",
       "  ['landuse', 101],\n",
       "  ['fractal', 91],\n",
       "  ['simulation', 22],\n",
       "  ['cities', 18],\n",
       "  ['gis', 17],\n",
       "  ['dynamics', 15],\n",
       "  ['agent', 13],\n",
       "  ['case', 11]],\n",
       " [['design digital fabrication', 10],\n",
       "  ['building information modelling', 3],\n",
       "  ['design process', 79],\n",
       "  ['digital fabrication', 70],\n",
       "  ['computational design', 60],\n",
       "  ['architectural design', 58],\n",
       "  ['design fabrication', 46],\n",
       "  ['robotic fabrication', 12],\n",
       "  ['form finding', 11],\n",
       "  ['fabrication', 285],\n",
       "  ['material', 249],\n",
       "  ['construction', 207],\n",
       "  ['tools', 145],\n",
       "  ['structures', 137],\n",
       "  ['processes', 126],\n",
       "  ['complex', 120],\n",
       "  ['structural', 117],\n",
       "  ['robotic', 23],\n",
       "  ['systems', 21],\n",
       "  ['performance', 19],\n",
       "  ['parametric', 18],\n",
       "  ['concrete', 15],\n",
       "  ['generative', 15],\n",
       "  ['physical', 15],\n",
       "  ['3d', 14],\n",
       "  ['manufacturing', 14],\n",
       "  ['scale', 14],\n",
       "  ['simulation', 12],\n",
       "  ['structure', 12],\n",
       "  ['adaptive', 11],\n",
       "  ['using', 11]],\n",
       " [['building information modeling', 5],\n",
       "  ['design process planning', 2],\n",
       "  ['bim adoption', 13],\n",
       "  ['case study', 12],\n",
       "  ['product data', 4],\n",
       "  ['construction', 84],\n",
       "  ['developed', 52],\n",
       "  ['planning', 48],\n",
       "  ['project', 43],\n",
       "  ['framework', 38],\n",
       "  ['time', 37],\n",
       "  ['systems', 13],\n",
       "  ['simulation', 11]],\n",
       " [['finite element analysis', 9],\n",
       "  ['algorithm automatic 2d', 2],\n",
       "  ['solid model', 29],\n",
       "  ['mesh generation', 27],\n",
       "  ['case study', 22],\n",
       "  ['product configuration', 21],\n",
       "  ['design process', 19],\n",
       "  ['design rationale', 10],\n",
       "  ['mesh', 178],\n",
       "  ['configuration', 129],\n",
       "  ['algorithm', 121],\n",
       "  ['knowledge', 112],\n",
       "  ['construction', 100],\n",
       "  ['product', 84],\n",
       "  ['systems', 84],\n",
       "  ['surface', 81],\n",
       "  ['meshing', 14],\n",
       "  ['3d', 12],\n",
       "  ['engineering', 12],\n",
       "  ['multi', 12],\n",
       "  ['feature', 11],\n",
       "  ['geometric', 11]],\n",
       " [['geometric constraint solving', 14],\n",
       "  ['geometric constraint systems', 6],\n",
       "  ['geometric constraint', 32],\n",
       "  ['point cloud', 29],\n",
       "  ['boolean operations', 27],\n",
       "  ['data structure', 23],\n",
       "  ['geometric constraints', 23],\n",
       "  ['surface reconstruction', 14],\n",
       "  ['geometric', 236],\n",
       "  ['surface', 180],\n",
       "  ['algorithm', 175],\n",
       "  ['set', 121],\n",
       "  ['constraints', 115],\n",
       "  ['representation', 113],\n",
       "  ['mesh', 111],\n",
       "  ['problem', 106],\n",
       "  ['tolerance', 99],\n",
       "  ['objects', 96],\n",
       "  ['geometry', 95],\n",
       "  ['cad', 20],\n",
       "  ['3d', 19],\n",
       "  ['solid', 19],\n",
       "  ['surfaces', 19],\n",
       "  ['boundary', 16],\n",
       "  ['parametric', 14],\n",
       "  ['reconstruction', 12],\n",
       "  ['feature', 12],\n",
       "  ['topological', 12],\n",
       "  ['systems', 11],\n",
       "  ['computer', 11],\n",
       "  ['decomposition', 11],\n",
       "  ['non', 11]],\n",
       " [['building information modeling', 106],\n",
       "  ['information modeling bim', 76],\n",
       "  ['industry foundation classes', 30],\n",
       "  ['architecture engineering construction', 24],\n",
       "  ['building information models', 22],\n",
       "  ['foundation classes ifc', 21],\n",
       "  ['semantic web technologies', 14],\n",
       "  ['building information model', 11],\n",
       "  ['bv building information', 11],\n",
       "  ['engineering construction aec', 11],\n",
       "  ['building information modelling', 5],\n",
       "  ['building information', 151],\n",
       "  ['construction industry', 44],\n",
       "  ['case study', 42],\n",
       "  ['information modeling', 32],\n",
       "  ['building design', 28],\n",
       "  ['bim based', 19],\n",
       "  ['ontology based', 13],\n",
       "  ['bim', 419],\n",
       "  ['construction', 308],\n",
       "  ['project', 200],\n",
       "  ['management', 173],\n",
       "  ['framework', 107],\n",
       "  ['tools', 104],\n",
       "  ['knowledge', 102],\n",
       "  ['ontology', 101],\n",
       "  ['developed', 98],\n",
       "  ['automated', 16],\n",
       "  ['performance', 16],\n",
       "  ['support', 16],\n",
       "  ['integrated', 15],\n",
       "  ['checking', 13],\n",
       "  ['representation', 13],\n",
       "  ['energy', 12],\n",
       "  ['integrating', 12],\n",
       "  ['safety', 12],\n",
       "  ['systems', 12],\n",
       "  ['collaborative', 11]],\n",
       " [['finite element analysis', 10],\n",
       "  ['free form surfaces', 7],\n",
       "  ['control points', 88],\n",
       "  ['bspline curve', 54],\n",
       "  ['data points', 48],\n",
       "  ['curves surfaces', 45],\n",
       "  ['bezier curves', 39],\n",
       "  ['spline curves', 21],\n",
       "  ['spline surfaces', 19],\n",
       "  ['nurbs surfaces', 14],\n",
       "  ['subdivision surfaces', 14],\n",
       "  ['spline surface', 13],\n",
       "  ['surface', 596],\n",
       "  ['surfaces', 454],\n",
       "  ['curves', 431],\n",
       "  ['curve', 344],\n",
       "  ['points', 269],\n",
       "  ['algorithm', 239],\n",
       "  ['geometric', 197],\n",
       "  ['bspline', 172],\n",
       "  ['given', 170],\n",
       "  ['presented', 147],\n",
       "  ['shape', 144],\n",
       "  ['rational', 140],\n",
       "  ['subdivision', 131],\n",
       "  ['spline', 81],\n",
       "  ['interpolation', 45],\n",
       "  ['splines', 43],\n",
       "  ['approximation', 34],\n",
       "  ['cubic', 29],\n",
       "  ['fitting', 28],\n",
       "  ['parametric', 25],\n",
       "  ['curvature', 24],\n",
       "  ['nurbs', 20],\n",
       "  ['using', 20],\n",
       "  ['offset', 18],\n",
       "  ['free', 15],\n",
       "  ['patches', 15],\n",
       "  ['bezier', 14],\n",
       "  ['blending', 14],\n",
       "  ['computing', 14],\n",
       "  ['freeform', 14],\n",
       "  ['planar', 14]],\n",
       " [['architectural design process', 11],\n",
       "  ['architectural design studio', 2],\n",
       "  ['design process', 143],\n",
       "  ['design processes', 49],\n",
       "  ['architectural design', 43],\n",
       "  ['design research', 33],\n",
       "  ['design thinking', 31],\n",
       "  ['conceptual design', 10],\n",
       "  ['designers', 310],\n",
       "  ['knowledge', 120],\n",
       "  ['students', 118],\n",
       "  ['cognitive', 113],\n",
       "  ['problem', 100],\n",
       "  ['understanding', 94],\n",
       "  ['sketching', 84],\n",
       "  ['studies', 84],\n",
       "  ['practice', 77],\n",
       "  ['engineering', 21],\n",
       "  ['designing', 19],\n",
       "  ['creative', 18],\n",
       "  ['systems', 17],\n",
       "  ['creativity', 16],\n",
       "  ['collaborative', 14],\n",
       "  ['learning', 13],\n",
       "  ['teaching', 12],\n",
       "  ['tools', 12],\n",
       "  ['cognition', 11],\n",
       "  ['parametric', 11],\n",
       "  ['product', 11],\n",
       "  ['user', 11],\n",
       "  ['visual', 11]],\n",
       " [['building information modeling', 64],\n",
       "  ['information modeling bim', 47],\n",
       "  ['tool path generation', 39],\n",
       "  ['support vector machine', 29],\n",
       "  ['computer aided design', 18],\n",
       "  ['genetic algorithm ga', 17],\n",
       "  ['particle swarm optimization', 16],\n",
       "  ['finite element analysis', 15],\n",
       "  ['experimental results demonstrate', 14],\n",
       "  ['finite element method', 14],\n",
       "  ['sculptured surface machining', 9],\n",
       "  ['case study', 186],\n",
       "  ['proposed method', 182],\n",
       "  ['design process', 111],\n",
       "  ['tool path', 103],\n",
       "  ['construction industry', 98],\n",
       "  ['real time', 28],\n",
       "  ['construction projects', 18],\n",
       "  ['path planning', 17],\n",
       "  ['augmented reality', 16],\n",
       "  ['based approach', 16],\n",
       "  ['construction', 1319],\n",
       "  ['algorithm', 777],\n",
       "  ['surface', 661],\n",
       "  ['developed', 626],\n",
       "  ['performance', 558],\n",
       "  ['simulation', 96],\n",
       "  ['3d', 84],\n",
       "  ['management', 77],\n",
       "  ['machining', 75],\n",
       "  ['optimization', 68],\n",
       "  ['automated', 68],\n",
       "  ['control', 68],\n",
       "  ['planning', 66],\n",
       "  ['multi', 64],\n",
       "  ['surfaces', 62],\n",
       "  ['axis', 56],\n",
       "  ['engineering', 52],\n",
       "  ['systems', 52],\n",
       "  ['evaluation', 51],\n",
       "  ['automatic', 50],\n",
       "  ['application', 48],\n",
       "  ['cad', 48],\n",
       "  ['integrated', 48]],\n",
       " [['human behavior simulation', 6],\n",
       "  ['classicism new synthesis', 2],\n",
       "  ['parametric design', 24],\n",
       "  ['architectural design', 17],\n",
       "  ['parametric modeling', 17],\n",
       "  ['design process', 14],\n",
       "  ['case studies', 13],\n",
       "  ['design space', 5],\n",
       "  ['parametric', 56],\n",
       "  ['systems', 50],\n",
       "  ['performance', 46],\n",
       "  ['tools', 46],\n",
       "  ['designers', 43],\n",
       "  ['planning', 17],\n",
       "  ['urban', 12]],\n",
       " [['building information modeling', 8],\n",
       "  ['land use planning', 8],\n",
       "  ['case study', 37],\n",
       "  ['design process', 28],\n",
       "  ['virtual reality', 23],\n",
       "  ['case studies', 22],\n",
       "  ['architectural design', 20],\n",
       "  ['computer aided', 8],\n",
       "  ['construction', 172],\n",
       "  ['planning', 151],\n",
       "  ['urban', 150],\n",
       "  ['project', 128],\n",
       "  ['problem', 123],\n",
       "  ['decision', 22],\n",
       "  ['systems', 21],\n",
       "  ['simulation', 20],\n",
       "  ['support', 20],\n",
       "  ['knowledge', 18],\n",
       "  ['environment', 17],\n",
       "  ['computer', 16],\n",
       "  ['evaluation', 16],\n",
       "  ['algorithm', 15],\n",
       "  ['optimization', 15],\n",
       "  ['application', 14],\n",
       "  ['environmental', 14],\n",
       "  ['new', 14],\n",
       "  ['automated', 13],\n",
       "  ['cad', 13],\n",
       "  ['making', 13]],\n",
       " [['analog digital notational', 2],\n",
       "  ['art net london', 2],\n",
       "  ['architectural design', 8],\n",
       "  ['modern architecture', 3],\n",
       "  ['urban', 44],\n",
       "  ['planning', 22],\n",
       "  ['physical', 21],\n",
       "  ['systems', 20],\n",
       "  ['cities', 16],\n",
       "  ['interactive', 16],\n",
       "  ['project', 16],\n",
       "  ['relationship', 16],\n",
       "  ['city', 15],\n",
       "  ['new', 10]],\n",
       " [['creative conceptual design', 9],\n",
       "  ['case based design', 8],\n",
       "  ['design process', 85],\n",
       "  ['conceptual design', 68],\n",
       "  ['engineering design', 45],\n",
       "  ['case study', 29],\n",
       "  ['design knowledge', 29],\n",
       "  ['case based', 13],\n",
       "  ['function', 198],\n",
       "  ['knowledge', 161],\n",
       "  ['product', 119],\n",
       "  ['systems', 119],\n",
       "  ['functional', 106],\n",
       "  ['engineering', 97],\n",
       "  ['representation', 91],\n",
       "  ['learning', 13],\n",
       "  ['planning', 13],\n",
       "  ['reasoning', 13],\n",
       "  ['urban', 13]]]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cluster: 0\n",
      "shape grammars (119x); shape grammar (110x); design process (34x); architectural design (33x); design space (20x); generative design (11x);\n",
      "shape (271x); grammar (158x); rules (140x); shapes (113x); grammars (91x); generation (74x); presented (73x); systems (70x); language (17x);\n",
      "plans (15x); implementation (14x); planning (14x); formal (13x); layout (13x); representation (13x); computer (12x); floor (12x); urban\n",
      "(12x); generative (11x)\n",
      "--------------------\n",
      "cluster: 1\n",
      "point cloud data (17x); radio frequency identification (16x); building information models (13x); precast concrete elements (12x); building\n",
      "information modeling (11x); construction sites (55x); construction site (54x); point clouds (42x); construction industry (35x); point cloud\n",
      "(32x); real time (15x); construction (491x); safety (156x); equipment (132x); tracking (117x); workers (106x); management (98x); project\n",
      "(93x); technology (87x); 3d (33x); automated (30x); laser (15x); monitoring (15x); bim (14x); control (14x); automatic (13x); rfid (13x);\n",
      "vision (12x); assessment (11x); detection (11x); performance (11x)\n",
      "--------------------\n",
      "cluster: 2\n",
      "space syntax (65x); street network (21x); spatial structure (20x); urban design (19x); spatial configuration (15x); urban (208x); patterns\n",
      "(85x); movement (68x); network (65x); properties (62x); measures (57x); networks (55x); morphology (11x)\n",
      "--------------------\n",
      "cluster: 3\n",
      "spatial planning (18x); planning process (13x); planning (281x); urban (63x); local (60x); plans (53x); performance (48x); public (46x);\n",
      "practice (43x); plan (42x); implementation (40x); planners (33x); case (12x)\n",
      "--------------------\n",
      "cluster: 4\n",
      "cellular automata ca (17x); urban growth (53x); cellular automata (29x); landuse change (28x); urban planning (26x); urban design (25x);\n",
      "planning support (17x); urban (437x); planning (214x); land (109x); city (105x); systems (104x); landuse (101x); fractal (91x); simulation\n",
      "(22x); cities (18x); gis (17x); dynamics (15x); agent (13x); case (11x)\n",
      "--------------------\n",
      "cluster: 5\n",
      "design process (79x); digital fabrication (70x); computational design (60x); architectural design (58x); design fabrication (46x); robotic\n",
      "fabrication (12x); form finding (11x); fabrication (285x); material (249x); construction (207x); tools (145x); structures (137x); processes\n",
      "(126x); complex (120x); structural (117x); robotic (23x); systems (21x); performance (19x); parametric (18x); concrete (15x); generative\n",
      "(15x); physical (15x); 3d (14x); manufacturing (14x); scale (14x); simulation (12x); structure (12x); adaptive (11x); using (11x)\n",
      "--------------------\n",
      "cluster: 6\n",
      "bim adoption (13x); case study (12x); construction (84x); developed (52x); planning (48x); project (43x); framework (38x); time (37x);\n",
      "systems (13x); simulation (11x)\n",
      "--------------------\n",
      "cluster: 7\n",
      "solid model (29x); mesh generation (27x); case study (22x); product configuration (21x); design process (19x); mesh (178x); configuration\n",
      "(129x); algorithm (121x); knowledge (112x); construction (100x); product (84x); systems (84x); surface (81x); meshing (14x); 3d (12x);\n",
      "engineering (12x); multi (12x); feature (11x); geometric (11x)\n",
      "--------------------\n",
      "cluster: 8\n",
      "geometric constraint solving (14x); geometric constraint (32x); point cloud (29x); boolean operations (27x); data structure (23x); geometric\n",
      "constraints (23x); surface reconstruction (14x); geometric (236x); surface (180x); algorithm (175x); set (121x); constraints (115x);\n",
      "representation (113x); mesh (111x); problem (106x); tolerance (99x); objects (96x); geometry (95x); cad (20x); 3d (19x); solid (19x);\n",
      "surfaces (19x); boundary (16x); parametric (14x); reconstruction (12x); feature (12x); topological (12x); systems (11x); computer (11x);\n",
      "decomposition (11x); non (11x)\n",
      "--------------------\n",
      "cluster: 9\n",
      "building information modeling (106x); information modeling bim (76x); industry foundation classes (30x); architecture engineering\n",
      "construction (24x); building information models (22x); foundation classes ifc (21x); semantic web technologies (14x); building information\n",
      "model (11x); bv building information (11x); engineering construction aec (11x); building information (151x); construction industry (44x);\n",
      "case study (42x); information modeling (32x); building design (28x); bim based (19x); ontology based (13x); bim (419x); construction (308x);\n",
      "project (200x); management (173x); framework (107x); tools (104x); knowledge (102x); ontology (101x); developed (98x); automated (16x);\n",
      "performance (16x); support (16x); integrated (15x); checking (13x); representation (13x); energy (12x); integrating (12x); safety (12x);\n",
      "systems (12x); collaborative (11x)\n",
      "--------------------\n",
      "cluster: 10\n",
      "control points (88x); bspline curve (54x); data points (48x); curves surfaces (45x); bezier curves (39x); spline curves (21x); spline\n",
      "surfaces (19x); nurbs surfaces (14x); subdivision surfaces (14x); spline surface (13x); surface (596x); surfaces (454x); curves (431x);\n",
      "curve (344x); points (269x); algorithm (239x); geometric (197x); bspline (172x); given (170x); presented (147x); shape (144x); rational\n",
      "(140x); subdivision (131x); spline (81x); interpolation (45x); splines (43x); approximation (34x); cubic (29x); fitting (28x); parametric\n",
      "(25x); curvature (24x); nurbs (20x); using (20x); offset (18x); free (15x); patches (15x); bezier (14x); blending (14x); computing (14x);\n",
      "freeform (14x); planar (14x)\n",
      "--------------------\n",
      "cluster: 11\n",
      "architectural design process (11x); design process (143x); design processes (49x); architectural design (43x); design research (33x); design\n",
      "thinking (31x); designers (310x); knowledge (120x); students (118x); cognitive (113x); problem (100x); understanding (94x); sketching (84x);\n",
      "studies (84x); practice (77x); engineering (21x); designing (19x); creative (18x); systems (17x); creativity (16x); collaborative (14x);\n",
      "learning (13x); teaching (12x); tools (12x); cognition (11x); parametric (11x); product (11x); user (11x); visual (11x)\n",
      "--------------------\n",
      "cluster: 12\n",
      "building information modeling (64x); information modeling bim (47x); tool path generation (39x); support vector machine (29x); computer\n",
      "aided design (18x); genetic algorithm ga (17x); particle swarm optimization (16x); finite element analysis (15x); experimental results\n",
      "demonstrate (14x); finite element method (14x); case study (186x); proposed method (182x); design process (111x); tool path (103x);\n",
      "construction industry (98x); real time (28x); construction projects (18x); path planning (17x); augmented reality (16x); based approach\n",
      "(16x); construction (1319x); algorithm (777x); surface (661x); developed (626x); performance (558x); simulation (96x); 3d (84x); management\n",
      "(77x); machining (75x); optimization (68x); automated (68x); control (68x); planning (66x); multi (64x); surfaces (62x); axis (56x);\n",
      "engineering (52x); systems (52x); evaluation (51x); automatic (50x); application (48x); cad (48x); integrated (48x)\n",
      "--------------------\n",
      "cluster: 13\n",
      "parametric design (24x); architectural design (17x); parametric modeling (17x); design process (14x); case studies (13x); parametric (56x);\n",
      "systems (50x); performance (46x); tools (46x); designers (43x); planning (17x); urban (12x)\n",
      "--------------------\n",
      "cluster: 14\n",
      "case study (37x); design process (28x); virtual reality (23x); case studies (22x); architectural design (20x); construction (172x); planning\n",
      "(151x); urban (150x); project (128x); problem (123x); decision (22x); systems (21x); simulation (20x); support (20x); knowledge (18x);\n",
      "environment (17x); computer (16x); evaluation (16x); algorithm (15x); optimization (15x); application (14x); environmental (14x); new (14x);\n",
      "automated (13x); cad (13x); making (13x)\n",
      "--------------------\n",
      "cluster: 15\n",
      "urban (44x); planning (22x); physical (21x); systems (20x); cities (16x); interactive (16x); project (16x); relationship (16x); city (15x)\n",
      "--------------------\n",
      "cluster: 16\n",
      "design process (85x); conceptual design (68x); engineering design (45x); case study (29x); design knowledge (29x); case based (13x);\n",
      "function (198x); knowledge (161x); product (119x); systems (119x); functional (106x); engineering (97x); representation (91x); learning\n",
      "(13x); planning (13x); reasoning (13x); urban (13x)\n",
      "--------------------\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>shape grammars (119x); shape grammar (110x); d...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>point cloud data (17x); radio frequency identi...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>space syntax (65x); street network (21x); spat...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>spatial planning (18x); planning process (13x)...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>cellular automata ca (17x); urban growth (53x)...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>design process (79x); digital fabrication (70x...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>bim adoption (13x); case study (12x); construc...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>solid model (29x); mesh generation (27x); case...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>geometric constraint solving (14x); geometric ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>building information modeling (106x); informat...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>control points (88x); bspline curve (54x); dat...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>architectural design process (11x); design pro...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>building information modeling (64x); informati...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>parametric design (24x); architectural design ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>case study (37x); design process (28x); virtua...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>urban (44x); planning (22x); physical (21x); s...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>design process (85x); conceptual design (68x);...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                    0\n",
       "0   shape grammars (119x); shape grammar (110x); d...\n",
       "1   point cloud data (17x); radio frequency identi...\n",
       "2   space syntax (65x); street network (21x); spat...\n",
       "3   spatial planning (18x); planning process (13x)...\n",
       "4   cellular automata ca (17x); urban growth (53x)...\n",
       "5   design process (79x); digital fabrication (70x...\n",
       "6   bim adoption (13x); case study (12x); construc...\n",
       "7   solid model (29x); mesh generation (27x); case...\n",
       "8   geometric constraint solving (14x); geometric ...\n",
       "9   building information modeling (106x); informat...\n",
       "10  control points (88x); bspline curve (54x); dat...\n",
       "11  architectural design process (11x); design pro...\n",
       "12  building information modeling (64x); informati...\n",
       "13  parametric design (24x); architectural design ...\n",
       "14  case study (37x); design process (28x); virtua...\n",
       "15  urban (44x); planning (22x); physical (21x); s...\n",
       "16  design process (85x); conceptual design (68x);..."
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#import textwrap\n",
    "allscores = addClusterScores(trigrams,bigrams)\n",
    "allscores = addClusterScores(allscores,filteredwords)\n",
    "display(allscores)\n",
    "clusterlabels = []\n",
    "clusterlabelsrows = []\n",
    "clusterlabels_textblock = []\n",
    "#clusterlabelsb = []\n",
    "counter = 0\n",
    "linelength = 140\n",
    "for x in range(0,len(trigrams)):\n",
    "    #clustertext = \"TRIGRAMS: \" + textwrap.fill(\"; \".join([ (w[0]+'('+str(w[1])+')') for w in trigrams[x] if w[1]>10 or trigrams[x].index(w)==0]),linelength) + '\\n'\n",
    "    #clustertext = clustertext + \"BIGRAMS: \" + textwrap.fill(\"; \".join([ (w[0]+'('+str(w[1])+')') for w in bigrams[x] if w[1]>10]),linelength)+ '\\n'\n",
    "    #clustertext = clustertext + \"WORDS: \" + textwrap.fill(\"; \".join([ (w[0]+'('+str(w[1])+')') for w in filteredwords[x] if w[1]>10]),linelength)\n",
    "    trigramslabel =  \"; \".join([ (w[0]+' ('+str(w[1])+'x)') for w in trigrams[x] if w[1]>10 ]) #and trigrams[x].index(w)<5])\n",
    "    bigramslabel = \"; \".join([ (w[0]+' ('+str(w[1])+'x)') for w in bigrams[x] if w[1]>10 ]) #and bigrams[x].index(w)<5])\n",
    "    wordslabel = \"; \".join([ (w[0]+' ('+str(w[1])+'x)') for w in filteredwords[x] if w[1]>10 ]) #and filteredwords[x].index(w)<20])\n",
    "    #clustertext = \"OFTEN USED TERMS: \" + trigramslabel + (\"; \" if len(trigramslabel)>0 else \"\") + bigramslabel + (\"; \" if len(bigramslabel)>0 else \"\") + wordslabel\n",
    "    clustertext = trigramslabel + (\"; \" if len(trigramslabel)>0 else \"\") + bigramslabel + (\"; \" if len(bigramslabel)>0 else \"\") + wordslabel\n",
    "    clusterlabels_textblock.append(clustertext)\n",
    "    clustertext = textwrap.fill(clustertext,linelength) \n",
    "    clusterlabels.append(clustertext)\n",
    "    #this is how many lines are in this cluster's label\n",
    "    clusterlabelsrows.append(len(clustertext.split('\\n')))\n",
    "    #clusterlabelsb.append(\"\")\n",
    "    print(\"cluster: \" + str(counter))\n",
    "    #print(textwrap.fill(clustertext,70))\n",
    "    print(clustertext)\n",
    "    print(\"--------------------\")\n",
    "    counter +=1\n",
    "display(pd.DataFrame(clusterlabels_textblock))\n",
    "#print(clusterlabels)\n",
    "#display(clusterlabels_textblock)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### What are the most cited papers in each cluster\n",
    "Next we find the most cited papers for each cluster. This will give us the seminal works that everyone working that field always refers to. For example in the Shape Grammars cluster we have the [The Palladian Grammar](https://journals.sagepub.com/doi/abs/10.1068/b050005) by Stiny and Mitchel from 1978."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 129,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TOP 5 CITATIONS\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>10</th>\n",
       "      <th>11</th>\n",
       "      <th>12</th>\n",
       "      <th>13</th>\n",
       "      <th>14</th>\n",
       "      <th>15</th>\n",
       "      <th>16</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>[STINY G, 1978, ENVIRON PLANN B, V5, P5, DOI 1...</td>\n",
       "      <td>[Tang PB, 2010, AUTOMAT CONSTR, V19, P829, DOI...</td>\n",
       "      <td>[Hillier B, 1984, SOCIAL LOGIC SPACE, 116]</td>\n",
       "      <td>[Alexander C., 1964, NOTES SYNTHESIS FORM, 20]</td>\n",
       "      <td>[Clarke KC, 1997, ENVIRON PLANN B, V24, P247, ...</td>\n",
       "      <td>[Gramazio Fabio, 2008, DIGITAL MAT ARCHITEC, 33]</td>\n",
       "      <td>[SMITH BM, 1983, IEEE COMPUT GRAPH, V3, P78, D...</td>\n",
       "      <td>[Mittal S., 1989, P 11 INT JOINT C ART, P1395,...</td>\n",
       "      <td>[Requicha A.A.G., 1982, ACM COMPUT SURV, V12, ...</td>\n",
       "      <td>[Eastman C, 2011, BIM HDB GUIDE BUILDI, 53]</td>\n",
       "      <td>[Piegl L., 1997, NURBS BOOK, 85]</td>\n",
       "      <td>[Schon D, 1983, REFLECTIVE PRACTITIO, 110]</td>\n",
       "      <td>[CHOI BK, 1998, SCULPTURED SURFACE M, 38]</td>\n",
       "      <td>[Woodbury R, 2010, ELEMENTS PARAMETRIC, 30]</td>\n",
       "      <td>[Keeney R. L., 1976, DECISIONS MULTIPLE O, 18]</td>\n",
       "      <td>[FOX Michael, 2009, INTERACTIVE ARCHITEC, 8]</td>\n",
       "      <td>[GERO JS, 1990, AI MAG, V11, P26, 48]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>[STINY G, 1980, ENVIRON PLANN B, V7, P343, DOI...</td>\n",
       "      <td>[Bosche F, 2010, ADV ENG INFORM, V24, P107, DO...</td>\n",
       "      <td>[BENEDIKT ML, 1979, ENVIRON PLANN B, V6, P47, ...</td>\n",
       "      <td>[Forester J., 1989, PLANNING FACE POWER, 20]</td>\n",
       "      <td>[WHITE R, 1993, ENVIRON PLANN A, V25, P1175, D...</td>\n",
       "      <td>[Kolarevic B, 2005, PERFORMATIVE ARCHITE, 28]</td>\n",
       "      <td>[Smith SD, 1999, J CONSTR ENG M ASCE, V125, P1...</td>\n",
       "      <td>[MITTAL S, 1990, PROCEEDINGS : EIGHTH NATIONAL...</td>\n",
       "      <td>[Mantyla M., 1988, INTRO SOLID MODELING, 32]</td>\n",
       "      <td>[Eastman C., 2008, BIM HDB GUIDE BUILDI, 52]</td>\n",
       "      <td>[Hoschek J, 1993, FUNDAMENTALS COMPUTE, 56]</td>\n",
       "      <td>[Dorst K, 2001, DESIGN STUDIES, V22, P425, DOI...</td>\n",
       "      <td>[PERSSON H, 1978, COMPUT AIDED DESIGN, V10, P1...</td>\n",
       "      <td>[Woodbury RF, 2006, AI EDAM, V20, P63, DOI 10....</td>\n",
       "      <td>[MILLER GA, 1956, PSYCHOL REV, V63, P81, DOI 1...</td>\n",
       "      <td>[Harvey D., 1989, CONDITION POSTMODERN, 6]</td>\n",
       "      <td>[Hirtz J, 2002, RES ENG DES, V13, P65, DOI 10....</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>[KONING H, 1981, ENVIRON PLANN B, V8, P295, DO...</td>\n",
       "      <td>[Teizer J, 2010, AUTOMAT CONSTR, V19, P630, DO...</td>\n",
       "      <td>[Turner A, 2001, ENVIRON PLANN B, V28, P103, D...</td>\n",
       "      <td>[Healey P, 1996, ENVIRON PLANN B, V23, P217, D...</td>\n",
       "      <td>[Batty M., 1994, FRACTAL CITIES GEOME, 32]</td>\n",
       "      <td>[Kolarevic Branko., 2003, ARCHITECTURE DIGITAL...</td>\n",
       "      <td>[Smith L, 2010, ACTION RES-LONDON, V8, P407, D...</td>\n",
       "      <td>[BLACKER TD, 1991, INT J NUMER METH ENG, V32, ...</td>\n",
       "      <td>[HOFFMANN CM, 1989, GEOMETRIC SOLID MODE, 28]</td>\n",
       "      <td>[GRUBER TR, 1993, KNOWL ACQUIS, V5, P199, DOI ...</td>\n",
       "      <td>[De. Boor C., 1978, PRACTICAL GUIDE SPLI, 54]</td>\n",
       "      <td>[Goel V, 1995, SKETCHES THOUGHT, 54]</td>\n",
       "      <td>[LONEY GC, 1987, COMPUT AIDED DESIGN, V19, P85...</td>\n",
       "      <td>[Davis D, 2013, THESIS, 8]</td>\n",
       "      <td>[Cohon J. L., 1978, MULTIOBJECTIVE PROGR, 12]</td>\n",
       "      <td>[Graham S., 1996, TELECOMMUNICATIONS C, 5]</td>\n",
       "      <td>[Stone RB, 2000, J MECH DESIGN, V122, P359, DO...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>[Stiny G., 1972, Information Processing 71 Pro...</td>\n",
       "      <td>[Song J, 2006, AUTOMAT CONSTR, V15, P166, DOI ...</td>\n",
       "      <td>[HILLIER B, 1993, ENVIRON PLANN B, V20, P29, D...</td>\n",
       "      <td>[ALEXANDER ER, 1989, ENVIRON PLANN B, V16, P12...</td>\n",
       "      <td>[Parker DC, 2003, ANN ASSOC AM GEOGR, V93, P31...</td>\n",
       "      <td>[Kolarevic B., 2008, MANUFACTURING MAT EF, 19]</td>\n",
       "      <td>[Shah J. J., 1994, ADV FEATURE BASED MA, 3]</td>\n",
       "      <td>[Sabin D, 1998, IEEE INTELL SYST APP, V13, P42...</td>\n",
       "      <td>[LIGHT R, 1982, COMPUT AIDED DESIGN, V14, P209...</td>\n",
       "      <td>[Eastman C, 2009, AUTOMAT CONSTR, V18, P1011, ...</td>\n",
       "      <td>[Faux I. D., 1979, COMPUTATIONAL GEOMET, 53]</td>\n",
       "      <td>[Cross N, 2007, BOARD INT RES DES, P1, DOI 10....</td>\n",
       "      <td>[SURESH K, 1994, J ENG IND-T ASME, V116, P253,...</td>\n",
       "      <td>[Newell A, 1972, HUMAN PROBLEM SOLVIN, 7]</td>\n",
       "      <td>[LYNCH K., 1960, IMAGE CITY, 11]</td>\n",
       "      <td>[Tafuri Manfredo, 1976, ARCHITECTURE UTOPIA, 5]</td>\n",
       "      <td>[Umeda Y, 1996, AI EDAM, V10, P275, DOI 10.101...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>[Stiny G. N., 2006, SHAPE TALKING SEEING, 48]</td>\n",
       "      <td>[Bosche F, 2008, AUTOMAT CONSTR, V17, P499, DO...</td>\n",
       "      <td>[Hillier B, 1996, SPACE IS MACHINE, 33]</td>\n",
       "      <td>[RITTEL HWJ, 1973, POLICY SCI, V4, P155, DOI 1...</td>\n",
       "      <td>[COUCLELIS H, 1985, ENVIRON PLANN A, V17, P585...</td>\n",
       "      <td>[Khoshnevis B, 2004, AUTOMAT CONSTR, V13, P5, ...</td>\n",
       "      <td>[CHANG TC, 1990, EXPERT PROCESS PLANN, 3]</td>\n",
       "      <td>[PRICE MA, 1995, INT J NUMER METH ENG, V38, P3...</td>\n",
       "      <td>[BOUMA W, 1995, COMPUT AIDED DESIGN, V27, P487...</td>\n",
       "      <td>[Eastman C.M., 1999, BUILDING PRODUCT MOD, 29]</td>\n",
       "      <td>[CATMULL E, 1978, COMPUT AIDED DESIGN, V10, P3...</td>\n",
       "      <td>[Cross N., 1996, ANAL DESIGN ACTIVITY, 44]</td>\n",
       "      <td>[Preparata F., 1985, COMPUTATIONAL GEOMET, 25]</td>\n",
       "      <td>[Aish R, 2005, LECT NOTES COMPUT SC, V3638, P1...</td>\n",
       "      <td>[KIRKPATRICK S, 1983, SCIENCE, V220, P671, DOI...</td>\n",
       "      <td>[Rossi A., 1966, ARCHITETTURA CITTA, 5]</td>\n",
       "      <td>[Otto K, 2001, PRODUCT DESIGN TECHN, 27]</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                  0   \\\n",
       "0  [STINY G, 1978, ENVIRON PLANN B, V5, P5, DOI 1...   \n",
       "1  [STINY G, 1980, ENVIRON PLANN B, V7, P343, DOI...   \n",
       "2  [KONING H, 1981, ENVIRON PLANN B, V8, P295, DO...   \n",
       "3  [Stiny G., 1972, Information Processing 71 Pro...   \n",
       "4      [Stiny G. N., 2006, SHAPE TALKING SEEING, 48]   \n",
       "\n",
       "                                                  1   \\\n",
       "0  [Tang PB, 2010, AUTOMAT CONSTR, V19, P829, DOI...   \n",
       "1  [Bosche F, 2010, ADV ENG INFORM, V24, P107, DO...   \n",
       "2  [Teizer J, 2010, AUTOMAT CONSTR, V19, P630, DO...   \n",
       "3  [Song J, 2006, AUTOMAT CONSTR, V15, P166, DOI ...   \n",
       "4  [Bosche F, 2008, AUTOMAT CONSTR, V17, P499, DO...   \n",
       "\n",
       "                                                  2   \\\n",
       "0         [Hillier B, 1984, SOCIAL LOGIC SPACE, 116]   \n",
       "1  [BENEDIKT ML, 1979, ENVIRON PLANN B, V6, P47, ...   \n",
       "2  [Turner A, 2001, ENVIRON PLANN B, V28, P103, D...   \n",
       "3  [HILLIER B, 1993, ENVIRON PLANN B, V20, P29, D...   \n",
       "4            [Hillier B, 1996, SPACE IS MACHINE, 33]   \n",
       "\n",
       "                                                  3   \\\n",
       "0     [Alexander C., 1964, NOTES SYNTHESIS FORM, 20]   \n",
       "1       [Forester J., 1989, PLANNING FACE POWER, 20]   \n",
       "2  [Healey P, 1996, ENVIRON PLANN B, V23, P217, D...   \n",
       "3  [ALEXANDER ER, 1989, ENVIRON PLANN B, V16, P12...   \n",
       "4  [RITTEL HWJ, 1973, POLICY SCI, V4, P155, DOI 1...   \n",
       "\n",
       "                                                  4   \\\n",
       "0  [Clarke KC, 1997, ENVIRON PLANN B, V24, P247, ...   \n",
       "1  [WHITE R, 1993, ENVIRON PLANN A, V25, P1175, D...   \n",
       "2         [Batty M., 1994, FRACTAL CITIES GEOME, 32]   \n",
       "3  [Parker DC, 2003, ANN ASSOC AM GEOGR, V93, P31...   \n",
       "4  [COUCLELIS H, 1985, ENVIRON PLANN A, V17, P585...   \n",
       "\n",
       "                                                  5   \\\n",
       "0   [Gramazio Fabio, 2008, DIGITAL MAT ARCHITEC, 33]   \n",
       "1      [Kolarevic B, 2005, PERFORMATIVE ARCHITE, 28]   \n",
       "2  [Kolarevic Branko., 2003, ARCHITECTURE DIGITAL...   \n",
       "3     [Kolarevic B., 2008, MANUFACTURING MAT EF, 19]   \n",
       "4  [Khoshnevis B, 2004, AUTOMAT CONSTR, V13, P5, ...   \n",
       "\n",
       "                                                  6   \\\n",
       "0  [SMITH BM, 1983, IEEE COMPUT GRAPH, V3, P78, D...   \n",
       "1  [Smith SD, 1999, J CONSTR ENG M ASCE, V125, P1...   \n",
       "2  [Smith L, 2010, ACTION RES-LONDON, V8, P407, D...   \n",
       "3        [Shah J. J., 1994, ADV FEATURE BASED MA, 3]   \n",
       "4          [CHANG TC, 1990, EXPERT PROCESS PLANN, 3]   \n",
       "\n",
       "                                                  7   \\\n",
       "0  [Mittal S., 1989, P 11 INT JOINT C ART, P1395,...   \n",
       "1  [MITTAL S, 1990, PROCEEDINGS : EIGHTH NATIONAL...   \n",
       "2  [BLACKER TD, 1991, INT J NUMER METH ENG, V32, ...   \n",
       "3  [Sabin D, 1998, IEEE INTELL SYST APP, V13, P42...   \n",
       "4  [PRICE MA, 1995, INT J NUMER METH ENG, V38, P3...   \n",
       "\n",
       "                                                  8   \\\n",
       "0  [Requicha A.A.G., 1982, ACM COMPUT SURV, V12, ...   \n",
       "1       [Mantyla M., 1988, INTRO SOLID MODELING, 32]   \n",
       "2      [HOFFMANN CM, 1989, GEOMETRIC SOLID MODE, 28]   \n",
       "3  [LIGHT R, 1982, COMPUT AIDED DESIGN, V14, P209...   \n",
       "4  [BOUMA W, 1995, COMPUT AIDED DESIGN, V27, P487...   \n",
       "\n",
       "                                                  9   \\\n",
       "0        [Eastman C, 2011, BIM HDB GUIDE BUILDI, 53]   \n",
       "1       [Eastman C., 2008, BIM HDB GUIDE BUILDI, 52]   \n",
       "2  [GRUBER TR, 1993, KNOWL ACQUIS, V5, P199, DOI ...   \n",
       "3  [Eastman C, 2009, AUTOMAT CONSTR, V18, P1011, ...   \n",
       "4     [Eastman C.M., 1999, BUILDING PRODUCT MOD, 29]   \n",
       "\n",
       "                                                  10  \\\n",
       "0                   [Piegl L., 1997, NURBS BOOK, 85]   \n",
       "1        [Hoschek J, 1993, FUNDAMENTALS COMPUTE, 56]   \n",
       "2      [De. Boor C., 1978, PRACTICAL GUIDE SPLI, 54]   \n",
       "3       [Faux I. D., 1979, COMPUTATIONAL GEOMET, 53]   \n",
       "4  [CATMULL E, 1978, COMPUT AIDED DESIGN, V10, P3...   \n",
       "\n",
       "                                                  11  \\\n",
       "0         [Schon D, 1983, REFLECTIVE PRACTITIO, 110]   \n",
       "1  [Dorst K, 2001, DESIGN STUDIES, V22, P425, DOI...   \n",
       "2               [Goel V, 1995, SKETCHES THOUGHT, 54]   \n",
       "3  [Cross N, 2007, BOARD INT RES DES, P1, DOI 10....   \n",
       "4         [Cross N., 1996, ANAL DESIGN ACTIVITY, 44]   \n",
       "\n",
       "                                                  12  \\\n",
       "0          [CHOI BK, 1998, SCULPTURED SURFACE M, 38]   \n",
       "1  [PERSSON H, 1978, COMPUT AIDED DESIGN, V10, P1...   \n",
       "2  [LONEY GC, 1987, COMPUT AIDED DESIGN, V19, P85...   \n",
       "3  [SURESH K, 1994, J ENG IND-T ASME, V116, P253,...   \n",
       "4     [Preparata F., 1985, COMPUTATIONAL GEOMET, 25]   \n",
       "\n",
       "                                                  13  \\\n",
       "0        [Woodbury R, 2010, ELEMENTS PARAMETRIC, 30]   \n",
       "1  [Woodbury RF, 2006, AI EDAM, V20, P63, DOI 10....   \n",
       "2                         [Davis D, 2013, THESIS, 8]   \n",
       "3          [Newell A, 1972, HUMAN PROBLEM SOLVIN, 7]   \n",
       "4  [Aish R, 2005, LECT NOTES COMPUT SC, V3638, P1...   \n",
       "\n",
       "                                                  14  \\\n",
       "0     [Keeney R. L., 1976, DECISIONS MULTIPLE O, 18]   \n",
       "1  [MILLER GA, 1956, PSYCHOL REV, V63, P81, DOI 1...   \n",
       "2      [Cohon J. L., 1978, MULTIOBJECTIVE PROGR, 12]   \n",
       "3                   [LYNCH K., 1960, IMAGE CITY, 11]   \n",
       "4  [KIRKPATRICK S, 1983, SCIENCE, V220, P671, DOI...   \n",
       "\n",
       "                                                15  \\\n",
       "0     [FOX Michael, 2009, INTERACTIVE ARCHITEC, 8]   \n",
       "1       [Harvey D., 1989, CONDITION POSTMODERN, 6]   \n",
       "2       [Graham S., 1996, TELECOMMUNICATIONS C, 5]   \n",
       "3  [Tafuri Manfredo, 1976, ARCHITECTURE UTOPIA, 5]   \n",
       "4          [Rossi A., 1966, ARCHITETTURA CITTA, 5]   \n",
       "\n",
       "                                                  16  \n",
       "0              [GERO JS, 1990, AI MAG, V11, P26, 48]  \n",
       "1  [Hirtz J, 2002, RES ENG DES, V13, P65, DOI 10....  \n",
       "2  [Stone RB, 2000, J MECH DESIGN, V122, P359, DO...  \n",
       "3  [Umeda Y, 1996, AI EDAM, V10, P275, DOI 10.101...  \n",
       "4           [Otto K, 2001, PRODUCT DESIGN TECHN, 27]  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#THIS WILL GET US THE TOP CITED PAPERS PER CLUSTER\n",
    "#fullstrsl = []\n",
    "citationsDict = [None]*clusternum #per cluster\n",
    "fullscoreTopCitations = []\n",
    "\n",
    "for x in range(0,clusternum):\n",
    "    \n",
    "    citations = list(drc.loc[drc['cluster'] == x]['citestring'])\n",
    "    citationsDict[x] = {}\n",
    "    citations = [item for sublist in citations for item in sublist]\n",
    "    for citation in citations:\n",
    "        if not citation in citationsDict[x]:\n",
    "            citationsDict[x][citation] = [citation,1]\n",
    "        else:\n",
    "            citationsDict[x][citation][1] += 1\n",
    "    \n",
    "    #\n",
    "    rankedCitations = []\n",
    "    for citation in citationsDict[x]:\n",
    "        rankedCitations.append([citationsDict[x][citation][0],citationsDict[x][citation][1]])\n",
    "   \n",
    "    sorted_scores = sorted(rankedCitations, key=lambda x: x[1], reverse=True)\n",
    "    myscores = sorted_scores[0:5]\n",
    "    \n",
    "    scorelist = []\n",
    "    for s in myscores:\n",
    "        if s[1] > 1:\n",
    "            scorelist.append( [s[0],s[1]]) # do the word and the how often it comes)\n",
    "            #scorelist.append(s[0]+ ' ('+str(s[1]) + ')') # do the word and the how often it comes)\n",
    "    fullscoreTopCitations.append(scorelist)\n",
    "\n",
    "\n",
    "    \n",
    "print(\"TOP 5 CITATIONS\")\n",
    "display(pd.DataFrame(fullscoreTopCitations).transpose().head())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Most cited authors in a cluster\n",
    "\n",
    "Besides the individual works it is also very revealing who defines and drives the research in each cluster if we get a list of the most cited authors. So let's do that next. Sometimes the same author is logged differently in the WOS database, for example \"Gramazio Fabio\" and \"Gramazio F\". So first we will unify the authors list and then count the most popular."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "metadata": {},
   "outputs": [],
   "source": [
    "#cases for merging A to author B\n",
    "#1. \"gramazio fabio\" and \"gramazio f\" or \"knight t\" and \"knight tw\"\n",
    "#if both have two words, word[0] is same, word[1] has same first letter, name them with the longer second word\n",
    "\n",
    "#2. \"simon ha\" and \"simon h a\"\n",
    "#if A has 3 words (with 2nd and 3rd single letter)and B has 2 words (2nd of 2 letter) or vice versa and word[0] is same and word[1] same first letter and word[2] of the one who has it is same letter as the other one's word[1]'s second letter\n",
    "\n",
    "#3. \"stiny g\" and \"stiny g n\"\n",
    "##if A has 3 words (with 2nd and 3rd single letter)and B has 2 words (2nd of 1 letter) or vice versa and word[0] is same and word[1] same first letter\n",
    "\n",
    "def mergeSameAuthors(authorscorelist):\n",
    "    newscorelist = []\n",
    "    mergedict = {}\n",
    "    for author in authorscorelist:\n",
    "        #print(author)\n",
    "        if not 'none' in author[0] and not \"\".join(author[0].split()) == '':\n",
    "            names = author[0].split()  \n",
    "            #print(names)\n",
    "            newscore = [author[0],author[1]]\n",
    "            merged = False\n",
    "            if not names[0] in mergedict:\n",
    "                mergedict[names[0]] = []\n",
    "                #print(newscore)\n",
    "                #mergedict[names[0]].append(newscore)\n",
    "            else:\n",
    "                for i in range(0, len(mergedict[names[0]])):\n",
    "                    existingnames = mergedict[names[0]][i][0].split()\n",
    "                    if len(names) >= 2 and len(existingnames) >= 2:\n",
    "                        #case 1. and 2. and 3. from above\n",
    "                        if names[1][0] == existingnames[1][0]:\n",
    "                            #print(\"merging: (\" + author[0] + \") into (\" + mergedict[names[0]][i][0]) + \")\"\n",
    "                            mergedict[names[0]][i][1] = mergedict[names[0]][i][1] + author[1]\n",
    "                            merged = True\n",
    "            if not merged:\n",
    "                #print(\"appending: \" + newscore[0])\n",
    "                mergedict[names[0]].append(newscore)               \n",
    "    for familynamelist in mergedict:\n",
    "        #print(familynamelist)\n",
    "        for author in mergedict[familynamelist]:\n",
    "            #print(author)\n",
    "            newscorelist.append(author)\n",
    "    return newscorelist"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 128,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TOP CITED AUTHORS\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>10</th>\n",
       "      <th>11</th>\n",
       "      <th>12</th>\n",
       "      <th>13</th>\n",
       "      <th>14</th>\n",
       "      <th>15</th>\n",
       "      <th>16</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>[stiny gn , 356]</td>\n",
       "      <td>[teizer j , 85]</td>\n",
       "      <td>[hillier bill , 176]</td>\n",
       "      <td>[alexander ch , 62]</td>\n",
       "      <td>[batty ma , 243]</td>\n",
       "      <td>[kolarevic branko , 113]</td>\n",
       "      <td>[smith r t , 25]</td>\n",
       "      <td>[lee j i , 51]</td>\n",
       "      <td>[requicha a a , 140]</td>\n",
       "      <td>[eastman c a , 239]</td>\n",
       "      <td>[piegl les , 236]</td>\n",
       "      <td>[schon d-a , 259]</td>\n",
       "      <td>[choi bi , 131]</td>\n",
       "      <td>[woodbury r f , 53]</td>\n",
       "      <td>[mitchell wb , 30]</td>\n",
       "      <td>[jencks ca , 25]</td>\n",
       "      <td>[gero john s , 124]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>[knight terry , 170]</td>\n",
       "      <td>[tang pf , 63]</td>\n",
       "      <td>[turner ak , 68]</td>\n",
       "      <td>[alexander eh , 59]</td>\n",
       "      <td>[white r n , 75]</td>\n",
       "      <td>[menges achim , 86]</td>\n",
       "      <td>[smith s l , 24]</td>\n",
       "      <td>[lee s j , 47]</td>\n",
       "      <td>[hoffmann ch , 104]</td>\n",
       "      <td>[gruber tb , 62]</td>\n",
       "      <td>[farin gerald e , 229]</td>\n",
       "      <td>[cross nigel , 232]</td>\n",
       "      <td>[lee s h , 125]</td>\n",
       "      <td>[simon h , 28]</td>\n",
       "      <td>[keeney ralph l , 28]</td>\n",
       "      <td>[rossi aldo , 22]</td>\n",
       "      <td>[pahl gerhard , 78]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>[flemming u , 110]</td>\n",
       "      <td>[bosche fn , 63]</td>\n",
       "      <td>[batty ma , 67]</td>\n",
       "      <td>[healey p , 59]</td>\n",
       "      <td>[couclelis h , 57]</td>\n",
       "      <td>[gramazio f , 69]</td>\n",
       "      <td>[smith b m , 16]</td>\n",
       "      <td>[blacker t d , 33]</td>\n",
       "      <td>[rossignac jarek r , 77]</td>\n",
       "      <td>[sacks rafael , 47]</td>\n",
       "      <td>[sederberg thomas w , 151]</td>\n",
       "      <td>[lawson bryan , 132]</td>\n",
       "      <td>[chen y g , 118]</td>\n",
       "      <td>[davis db , 21]</td>\n",
       "      <td>[foley ja , 26]</td>\n",
       "      <td>[tafuri m , 22]</td>\n",
       "      <td>[goel ashok k , 73]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>[march lj , 104]</td>\n",
       "      <td>[brilakis ik , 56]</td>\n",
       "      <td>[peponis john , 65]</td>\n",
       "      <td>[faludi andreas , 49]</td>\n",
       "      <td>[clarke k , 56]</td>\n",
       "      <td>[oxman r e , 59]</td>\n",
       "      <td>[smith g c , 16]</td>\n",
       "      <td>[lee k h , 29]</td>\n",
       "      <td>[hoppe hugues , 61]</td>\n",
       "      <td>[azhar salman , 43]</td>\n",
       "      <td>[hoschek j , 123]</td>\n",
       "      <td>[dorst kee , 117]</td>\n",
       "      <td>[wang wenping , 115]</td>\n",
       "      <td>[eastman c a , 19]</td>\n",
       "      <td>[saaty thomas l , 26]</td>\n",
       "      <td>[harvey david , 18]</td>\n",
       "      <td>[brown d m , 69]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>[mitchell wb , 91]</td>\n",
       "      <td>[golparvar-fard mg , 53]</td>\n",
       "      <td>[benedikt m , 56]</td>\n",
       "      <td>[forester j , 44]</td>\n",
       "      <td>[wu fj , 55]</td>\n",
       "      <td>[hensel mu , 52]</td>\n",
       "      <td>[smith d m , 15]</td>\n",
       "      <td>[lee c h , 26]</td>\n",
       "      <td>[mantyla mj , 53]</td>\n",
       "      <td>[lee g s , 39]</td>\n",
       "      <td>[farouki ramt , 111]</td>\n",
       "      <td>[goldschmidt gabriela , 116]</td>\n",
       "      <td>[lee j -j , 111]</td>\n",
       "      <td>[akin o z , 17]</td>\n",
       "      <td>[simon hd , 26]</td>\n",
       "      <td>[alberti l , 16]</td>\n",
       "      <td>[maher m , 68]</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     0                         1                     2   \\\n",
       "0      [stiny gn , 356]           [teizer j , 85]  [hillier bill , 176]   \n",
       "1  [knight terry , 170]            [tang pf , 63]      [turner ak , 68]   \n",
       "2    [flemming u , 110]          [bosche fn , 63]       [batty ma , 67]   \n",
       "3      [march lj , 104]        [brilakis ik , 56]   [peponis john , 65]   \n",
       "4    [mitchell wb , 91]  [golparvar-fard mg , 53]     [benedikt m , 56]   \n",
       "\n",
       "                      3                   4                         5   \\\n",
       "0    [alexander ch , 62]    [batty ma , 243]  [kolarevic branko , 113]   \n",
       "1    [alexander eh , 59]    [white r n , 75]       [menges achim , 86]   \n",
       "2        [healey p , 59]  [couclelis h , 57]         [gramazio f , 69]   \n",
       "3  [faludi andreas , 49]     [clarke k , 56]          [oxman r e , 59]   \n",
       "4      [forester j , 44]        [wu fj , 55]          [hensel mu , 52]   \n",
       "\n",
       "                 6                   7                         8   \\\n",
       "0  [smith r t , 25]      [lee j i , 51]      [requicha a a , 140]   \n",
       "1  [smith s l , 24]      [lee s j , 47]       [hoffmann ch , 104]   \n",
       "2  [smith b m , 16]  [blacker t d , 33]  [rossignac jarek r , 77]   \n",
       "3  [smith g c , 16]      [lee k h , 29]       [hoppe hugues , 61]   \n",
       "4  [smith d m , 15]      [lee c h , 26]         [mantyla mj , 53]   \n",
       "\n",
       "                    9                           10  \\\n",
       "0  [eastman c a , 239]           [piegl les , 236]   \n",
       "1     [gruber tb , 62]      [farin gerald e , 229]   \n",
       "2  [sacks rafael , 47]  [sederberg thomas w , 151]   \n",
       "3  [azhar salman , 43]           [hoschek j , 123]   \n",
       "4       [lee g s , 39]        [farouki ramt , 111]   \n",
       "\n",
       "                             11                    12                   13  \\\n",
       "0             [schon d-a , 259]       [choi bi , 131]  [woodbury r f , 53]   \n",
       "1           [cross nigel , 232]       [lee s h , 125]       [simon h , 28]   \n",
       "2          [lawson bryan , 132]      [chen y g , 118]      [davis db , 21]   \n",
       "3             [dorst kee , 117]  [wang wenping , 115]   [eastman c a , 19]   \n",
       "4  [goldschmidt gabriela , 116]      [lee j -j , 111]      [akin o z , 17]   \n",
       "\n",
       "                      14                   15                   16  \n",
       "0     [mitchell wb , 30]     [jencks ca , 25]  [gero john s , 124]  \n",
       "1  [keeney ralph l , 28]    [rossi aldo , 22]  [pahl gerhard , 78]  \n",
       "2        [foley ja , 26]      [tafuri m , 22]  [goel ashok k , 73]  \n",
       "3  [saaty thomas l , 26]  [harvey david , 18]     [brown d m , 69]  \n",
       "4        [simon hd , 26]     [alberti l , 16]       [maher m , 68]  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fullstrsl = []\n",
    "for x in range(0,clusternum):\n",
    "    authors = list(drc.loc[drc['cluster'] == x]['citedAU'])\n",
    "    \n",
    "    authors = [item for sublist in authors for item in sublist]\n",
    "    #authors = authors + list(drc.loc[drc['cluster'] == x]['author'])\n",
    "    authors = \" §\".join(str(x) for x in authors)\n",
    "    #print(authors[:50])\n",
    "    #changed here from 2 to 0 because it otherwise combines different authors with same family name as one\n",
    "    #especially important for chinese names where family name such as \"wang\" is very common\n",
    "    #two thing can be done here\n",
    "    #1. merge entries for same author when he is listed with full name and family plus initial\n",
    "    #     such as \"gramazio fabio\" and \"gramazio f\"\n",
    "    #2. merge entries for same author when he is listed with first initial joined and not\n",
    "    #     such as \"simon ha\" and \"simon h a\"\n",
    "    #3. (optional as it might be wrong to do it) merge entries when author is listed with one and then with two initial\n",
    "    #     such as \"stiny g\" and \"stiny g n\"\n",
    "    #this might be done as follows\n",
    "    #authors = ' '.join( [w for w in authors.split() if len(w)>2] )\n",
    "    authors = ' '.join( [w for w in authors.split()])# if len(w)>0] )\n",
    "    #print(authors[:50])\n",
    "    fullstrsl.append(authors)\n",
    "\n",
    "#print(fullstrsl[1])\n",
    "vec = CountVectorizer(token_pattern=r'[\\s\\w\\.-]+(?=[$|§])')#Choose CountVectorizer for the most common words in the cluster, TfidfVectorizer for the words with the greatest differentiation value.\n",
    "X = vec.fit_transform(fullstrsl)\n",
    "\n",
    "clusterfeatures = pd.DataFrame(X.toarray(), columns=vec.get_feature_names())\n",
    "fullscoreA = []\n",
    "for x in range(0,clusternum):\n",
    "    scores = zip(vec.get_feature_names(), np.asarray(X[x,:].sum(axis=0)).ravel())\n",
    "    sorted_scores = sorted(scores, key=lambda x: x[1])\n",
    "    #print(sorted_scores[0:6])\n",
    "    sorted_scores = mergeSameAuthors(sorted_scores)\n",
    "    #print(sorted_scores[0:6])\n",
    "    sorted_scores = sorted(sorted_scores, key=lambda x: x[1], reverse=True)\n",
    "    myscores = sorted_scores[0:10]\n",
    "    \n",
    "    scorelist = []\n",
    "    for s in myscores:\n",
    "        if s[1] > 1:\n",
    "            scorelist.append( [s[0],s[1]]) # do the word and the how often it comes)\n",
    "            #scorelist.append(s[0]+ ' ('+str(s[1]) + ')') # do the word and the how often it comes)\n",
    "    fullscoreA.append(scorelist)\n",
    "print(\"TOP 5 CITED AUTHORS\")\n",
    "display(pd.DataFrame(fullscoreA).transpose().head())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Great, now we have a list of authors for each cluster. Let's also count the number of articles per cluster"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "9779\n",
      "415\n",
      "191\n",
      "214\n",
      "175\n",
      "322\n",
      "409\n",
      "176\n",
      "332\n",
      "351\n",
      "283\n",
      "549\n",
      "372\n",
      "1851\n",
      "211\n",
      "623\n",
      "175\n",
      "312\n"
     ]
    }
   ],
   "source": [
    "#this counts how many articles are in each cluster\n",
    "articleCount = []\n",
    "totalArticleCount = len(drc)\n",
    "print(totalArticleCount)\n",
    "for x in range(0,clusternum):\n",
    "    articles = len(drc.loc[drc['cluster'] == x])\n",
    "    articleCount.append(articles)\n",
    "    print(articles)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's assemble the text for the sublabels for each cluster from the most cited works, most cited authors and the number of articles."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 290,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cluster: 0\n",
      "MOST CITED AUTHORS: STINY,G. (356x); KNIGHT,T. (170x); FLEMMING,U. (110x); MARCH,L. (104x); MITCHELL,W. (91x);\n",
      "DUARTE,J. (86x); KRISHNAMURTI,R. (86x); GERO,J. (60x); COYNE,R. (59x); KONING,H. (57x)\n",
      "TOP CITATIONS: STINY G, 1978, ENVIRON PLANN B, V5, P5, DOI 10.1068/B050005 (107x); STINY G, 1980, ENVIRON\n",
      "PLANN B, V7, P343, DOI 10.1068/B070343 (92x); KONING H, 1981, ENVIRON PLANN B, V8, P295, DOI 10.1068/B080295\n",
      "(54x); STINY G., 1972, INFORMATION PROCESSING 71 PROCEEDINGS OF THE IFIP CONGRESS 1971. VOLUME 2, P1460 (53x);\n",
      "STINY G. N., 2006, SHAPE TALKING SEEING (48x)\n",
      "--------------------\n",
      "cluster: 1\n",
      "MOST CITED AUTHORS: TEIZER,J. (85x); TANG,P. (63x); BOSCHE,F. (63x); BRILAKIS,I. (56x); GOLPARVAR-FARD,M.\n",
      "(53x); SONG,J. (53x); CHENG,T. (52x); NAVON,R. (48x); AKINCI,B. (41x); ERGEN,E. (40x)\n",
      "TOP CITATIONS: TANG PB, 2010, AUTOMAT CONSTR, V19, P829, DOI 10.1016/J.AUTCON.2010.06.007 (38x); BOSCHE F,\n",
      "2010, ADV ENG INFORM, V24, P107, DOI 10.1016/J.AEI.2009.08.006 (33x); TEIZER J, 2010, AUTOMAT CONSTR, V19,\n",
      "P630, DOI 10.1016/J.AUTCON.2010.02.009 (30x); SONG J, 2006, AUTOMAT CONSTR, V15, P166, DOI\n",
      "10.1016/J.AUTCON.2005.03.001 (28x); BOSCHE F, 2008, AUTOMAT CONSTR, V17, P499, DOI\n",
      "10.1016/J.AUTCON.2007.09.001 (28x)\n",
      "--------------------\n",
      "cluster: 2\n",
      "MOST CITED AUTHORS: HILLIER,B. (176x); TURNER,A. (68x); BATTY,M. (67x); PEPONIS,J. (65x); BENEDIKT,M. (56x);\n",
      "LYNCH,K. (40x); PENN,A. (35x); HANSON,J. (32x); RATTI,C. (31x); STEADMAN,P. (29x)\n",
      "TOP CITATIONS: HILLIER B, 1984, SOCIAL LOGIC SPACE (116x); BENEDIKT ML, 1979, ENVIRON PLANN B, V6, P47, DOI\n",
      "10.1068/B060047 (51x); TURNER A, 2001, ENVIRON PLANN B, V28, P103, DOI 10.1068/B2684 (46x); HILLIER B, 1993,\n",
      "ENVIRON PLANN B, V20, P29, DOI 10.1068/B200029 (38x); HILLIER B, 1996, SPACE IS MACHINE (33x)\n",
      "--------------------\n",
      "cluster: 3\n",
      "MOST CITED AUTHORS: ALEXANDER,C. (62x); ALEXANDER,E. (59x); HEALEY,P. (59x); FALUDI,A. (49x); FORESTER,J.\n",
      "(44x); INNES,J. (41x); FRIEND,J. (34x); FRIEDMANN,J. (29x); BERKE,P. (23x); LINDBLOM,C. (23x)\n",
      "TOP CITATIONS: ALEXANDER C., 1964, NOTES SYNTHESIS FORM (20x); FORESTER J., 1989, PLANNING FACE POWER (20x);\n",
      "HEALEY P, 1996, ENVIRON PLANN B, V23, P217, DOI 10.1068/B230217 (14x); ALEXANDER ER, 1989, ENVIRON PLANN B,\n",
      "V16, P127, DOI 10.1068/B160127 (14x); RITTEL HWJ, 1973, POLICY SCI, V4, P155, DOI 10.1007/BF01405730 (13x)\n",
      "--------------------\n",
      "cluster: 4\n",
      "MOST CITED AUTHORS: BATTY,M. (243x); WHITE,R. (75x); COUCLELIS,H. (57x); CLARKE,K. (56x); WU,F. (55x);\n",
      "KLOSTERMAN,R. (49x); LANDIS,J. (48x); MANDELBROT,B. (48x); TORRENS,P. (44x); WILSON,A. (43x)\n",
      "TOP CITATIONS: CLARKE KC, 1997, ENVIRON PLANN B, V24, P247, DOI 10.1068/B240247 (39x); WHITE R, 1993, ENVIRON\n",
      "PLANN A, V25, P1175, DOI 10.1068/A251175 (34x); BATTY M., 1994, FRACTAL CITIES GEOME (32x); PARKER DC, 2003,\n",
      "ANN ASSOC AM GEOGR, V93, P314, DOI 10.1111/1467-8306.9302004 (25x); COUCLELIS H, 1985, ENVIRON PLANN A, V17,\n",
      "P585, DOI 10.1068/A170585 (24x)\n",
      "--------------------\n",
      "cluster: 5\n",
      "MOST CITED AUTHORS: KOLAREVIC,B. (113x); MENGES,A. (86x); GRAMAZIO,F. (69x); OXMAN,R. (59x); HENSEL,M. (52x);\n",
      "OXMAN,N. (50x); BURRY,M. (40x); RIPPMANN,M. (36x); BLOCK,P. (29x); FALLACARA,G. (28x)\n",
      "TOP CITATIONS: GRAMAZIO FABIO, 2008, DIGITAL MAT ARCHITEC (33x); KOLAREVIC B, 2005, PERFORMATIVE ARCHITE\n",
      "(28x); KOLAREVIC BRANKO., 2003, ARCHITECTURE DIGITAL (28x); KOLAREVIC B., 2008, MANUFACTURING MAT EF (19x);\n",
      "KHOSHNEVIS B, 2004, AUTOMAT CONSTR, V13, P5, DOI 10.1016/J.AUTCON.2003.08.012 (19x)\n",
      "--------------------\n",
      "cluster: 6\n",
      "MOST CITED AUTHORS: SMITH,R. (25x); SMITH,S. (24x); SMITH,B. (16x); SMITH,G. (16x); SMITH,D. (15x); SHAH,J.\n",
      "(12x); SMITH,A. (12x); SMITH,T. (12x); SMITH,J. (9x); SMITH,L. (9x)\n",
      "TOP CITATIONS: SMITH BM, 1983, IEEE COMPUT GRAPH, V3, P78, DOI 10.1109/MCG.1983.263328 (4x); SMITH SD, 1999, J\n",
      "CONSTR ENG M ASCE, V125, P133, DOI 10.1061/(ASCE)0733-9364(1999)125:3(133) (3x); SMITH L, 2010, ACTION RES-\n",
      "LONDON, V8, P407, DOI 10.1177/1476750310366043 (3x); SHAH J. J., 1994, ADV FEATURE BASED MA (3x); CHANG TC,\n",
      "1990, EXPERT PROCESS PLANN (3x)\n",
      "--------------------\n",
      "cluster: 7\n",
      "MOST CITED AUTHORS: LEE,J. (51x); LEE,S. (47x); BLACKER,T. (33x); LEE,K. (29x); LEE,C. (26x); OWEN,S. (26x);\n",
      "LEE,Y. (23x); GEORGE,P. (22x); MITTAL,S. (22x); SHEWCHUK,J. (20x)\n",
      "TOP CITATIONS: MITTAL S., 1989, P 11 INT JOINT C ART, P1395 (14x); MITTAL S, 1990, PROCEEDINGS : EIGHTH\n",
      "NATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE, VOLS 1 AND 2, P25 (13x); BLACKER TD, 1991, INT J NUMER METH\n",
      "ENG, V32, P811, DOI 10.1002/NME.1620320410 (12x); SABIN D, 1998, IEEE INTELL SYST APP, V13, P42, DOI\n",
      "10.1109/5254.708432 (11x); PRICE MA, 1995, INT J NUMER METH ENG, V38, P3335, DOI 10.1002/NME.1620381910 (10x)\n",
      "--------------------\n",
      "cluster: 8\n",
      "MOST CITED AUTHORS: REQUICHA,A. (140x); HOFFMANN,C. (104x); ROSSIGNAC,J. (77x); HOPPE,H. (61x); MANTYLA,M.\n",
      "(53x); AMENTA,N. (47x); DEY,T. (44x); LIGHT,R. (44x); OWEN,J. (43x); TAUBIN,G. (36x)\n",
      "TOP CITATIONS: REQUICHA A.A.G., 1982, ACM COMPUT SURV, V12, P437, DOI 10.1145/356827.356833 (58x); MANTYLA M.,\n",
      "1988, INTRO SOLID MODELING (32x); HOFFMANN CM, 1989, GEOMETRIC SOLID MODE (28x); LIGHT R, 1982, COMPUT AIDED\n",
      "DESIGN, V14, P209, DOI 10.1016/0010-4485(82)90292-5 (28x); BOUMA W, 1995, COMPUT AIDED DESIGN, V27, P487, DOI\n",
      "10.1016/0010-4485(94)00013-4 (26x)\n",
      "--------------------\n",
      "cluster: 9\n",
      "MOST CITED AUTHORS: EASTMAN,C. (239x); GRUBER,T. (62x); SACKS,R. (47x); AZHAR,S. (43x); LEE,G. (39x); ZHANG,S.\n",
      "(35x); BEETZ,J. (34x); PAUWELS,P. (33x); SUCCAR,B. (32x); LIEBICH,T. (30x)\n",
      "TOP CITATIONS: EASTMAN C, 2011, BIM HDB GUIDE BUILDI (53x); EASTMAN C., 2008, BIM HDB GUIDE BUILDI (52x);\n",
      "GRUBER TR, 1993, KNOWL ACQUIS, V5, P199, DOI 10.1006/KNAC.1993.1008 (35x); EASTMAN C, 2009, AUTOMAT CONSTR,\n",
      "V18, P1011, DOI 10.1016/J.AUTCON.2009.07.002 (33x); EASTMAN C.M., 1999, BUILDING PRODUCT MOD (29x)\n",
      "--------------------\n",
      "cluster: 10\n",
      "MOST CITED AUTHORS: PIEGL,L. (236x); FARIN,G. (229x); SEDERBERG,T. (151x); HOSCHEK,J. (123x); FAROUKI,R.\n",
      "(111x); DE,B. (96x); FAUX,I. (77x); BOEHM,W. (75x); ELBER,G. (68x); TILLER,W. (60x)\n",
      "TOP CITATIONS: PIEGL L., 1997, NURBS BOOK (85x); HOSCHEK J, 1993, FUNDAMENTALS COMPUTE (56x); DE. BOOR C.,\n",
      "1978, PRACTICAL GUIDE SPLI (54x); FAUX I. D., 1979, COMPUTATIONAL GEOMET (53x); CATMULL E, 1978, COMPUT AIDED\n",
      "DESIGN, V10, P350, DOI 10.1016/0010-4485(78)90110-0 (46x)\n",
      "--------------------\n",
      "cluster: 11\n",
      "MOST CITED AUTHORS: SCHON,D. (259x); CROSS,N. (232x); LAWSON,B. (132x); DORST,K. (117x); GOLDSCHMIDT,G.\n",
      "(116x); GERO,J. (102x); SIMON,H. (101x); GOEL,V. (79x); SUWA,M. (76x); ERICSSON,K. (65x)\n",
      "TOP CITATIONS: SCHON D, 1983, REFLECTIVE PRACTITIO (110x); DORST K, 2001, DESIGN STUDIES, V22, P425, DOI DOI\n",
      "10.1016/S0142-694X(01)00009-6 (58x); GOEL V, 1995, SKETCHES THOUGHT (54x); CROSS N, 2007, BOARD INT RES DES,\n",
      "P1, DOI 10.1007/978-3-7643-8485-2 (45x); CROSS N., 1996, ANAL DESIGN ACTIVITY (44x)\n",
      "--------------------\n",
      "cluster: 12\n",
      "MOST CITED AUTHORS: CHOI,B. (131x); LEE,S. (125x); CHEN,Y. (118x); WANG,W. (115x); LEE,J. (111x); LIU,Y.\n",
      "(108x); WANG,J. (107x); WANG,X. (103x); WANG,Y. (101x); CHEN,J. (99x)\n",
      "TOP CITATIONS: CHOI BK, 1998, SCULPTURED SURFACE M (38x); PERSSON H, 1978, COMPUT AIDED DESIGN, V10, P169, DOI\n",
      "10.1016/0010-4485(78)90141-0 (29x); LONEY GC, 1987, COMPUT AIDED DESIGN, V19, P85, DOI\n",
      "10.1016/S0010-4485(87)80050-7 (27x); SURESH K, 1994, J ENG IND-T ASME, V116, P253, DOI 10.1115/1.2901938\n",
      "(25x); PREPARATA F., 1985, COMPUTATIONAL GEOMET (25x)\n",
      "--------------------\n",
      "cluster: 13\n",
      "MOST CITED AUTHORS: WOODBURY,R. (53x); SIMON,H. (28x); DAVIS,D. (21x); EASTMAN,C. (19x); AKIN,O. (17x);\n",
      "AISH,R. (16x); DAVIS,M. (14x); OXMAN,R. (13x); STINY,G. (11x); JOHNSON,S. (10x)\n",
      "TOP CITATIONS: WOODBURY R, 2010, ELEMENTS PARAMETRIC (30x); WOODBURY RF, 2006, AI EDAM, V20, P63, DOI\n",
      "10.1017/S0890060406060057 (9x); DAVIS D, 2013, THESIS (8x); NEWELL A, 1972, HUMAN PROBLEM SOLVIN (7x); AISH R,\n",
      "2005, LECT NOTES COMPUT SC, V3638, P151 (7x)\n",
      "--------------------\n",
      "cluster: 14\n",
      "MOST CITED AUTHORS: MITCHELL,W. (30x); KEENEY,R. (28x); FOLEY,J. (26x); SAATY,T. (26x); SIMON,H. (26x);\n",
      "MILLER,G. (25x); CHURCH,R. (24x); SUTHERLAND,I. (20x); WILSON,A. (20x); HOPKINS,L. (19x)\n",
      "TOP CITATIONS: KEENEY R. L., 1976, DECISIONS MULTIPLE O (18x); MILLER GA, 1956, PSYCHOL REV, V63, P81, DOI\n",
      "10.1037/0033-295X.101.2.343 (13x); COHON J. L., 1978, MULTIOBJECTIVE PROGR (12x); LYNCH K., 1960, IMAGE CITY\n",
      "(11x); KIRKPATRICK S, 1983, SCIENCE, V220, P671, DOI 10.1126/SCIENCE.220.4598.671 (10x)\n",
      "--------------------\n",
      "cluster: 15\n",
      "MOST CITED AUTHORS: JENCKS,C. (25x); ROSSI,A. (22x); TAFURI,M. (22x); HARVEY,D. (18x); ALBERTI,L. (16x);\n",
      "ROWE,C. (15x); FRAMPTON,K. (14x); BANHAM,R. (12x); EISENMAN,P. (12x); FOX,M. (12x)\n",
      "TOP CITATIONS: FOX MICHAEL, 2009, INTERACTIVE ARCHITEC (8x); HARVEY D., 1989, CONDITION POSTMODERN (6x);\n",
      "GRAHAM S., 1996, TELECOMMUNICATIONS C (5x); TAFURI MANFREDO, 1976, ARCHITECTURE UTOPIA (5x); ROSSI A., 1966,\n",
      "ARCHITETTURA CITTA (5x)\n",
      "--------------------\n",
      "cluster: 16\n",
      "MOST CITED AUTHORS: GERO,J. (124x); PAHL,G. (78x); GOEL,A. (73x); BROWN,D. (69x); MAHER,M. (68x);\n",
      "CHANDRASEKARAN,B. (57x); UMEDA,Y. (56x); STONE,R. (54x); CHAKRABARTI,A. (46x); ULRICH,K. (45x)\n",
      "TOP CITATIONS: GERO JS, 1990, AI MAG, V11, P26 (48x); HIRTZ J, 2002, RES ENG DES, V13, P65, DOI\n",
      "10.1007/S00163-001-0008K-3 (35x); STONE RB, 2000, J MECH DESIGN, V122, P359, DOI 10.1115/1.1289637 (35x);\n",
      "UMEDA Y, 1996, AI EDAM, V10, P275, DOI 10.1017/S0890060400001621 (33x); OTTO K, 2001, PRODUCT DESIGN TECHN\n",
      "(27x)\n",
      "--------------------\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "['STINY,GN (356x)\\nKNIGHT,TERRY (170x)\\nFLEMMING,U (110x)\\nMARCH,LJ (104x)\\nMITCHELL,WB (91x)\\nDUARTE,JS (86x)\\nKRISHNAMURTI,RAMESH (86x)\\nGERO,JOHN (60x)\\nCOYNE,RDD (59x)\\nKONING,H (57x)',\n",
       " 'TEIZER,J (85x)\\nTANG,PF (63x)\\nBOSCHE,FN (63x)\\nBRILAKIS,IK (56x)\\nGOLPARVAR-FARD,MG (53x)\\nSONG,JB (53x)\\nCHENG,T (52x)\\nNAVON,R (48x)\\nAKINCI,BURCU (41x)\\nERGEN,E (40x)',\n",
       " 'HILLIER,BILL (176x)\\nTURNER,AK (68x)\\nBATTY,MA (67x)\\nPEPONIS,JOHN (65x)\\nBENEDIKT,M (56x)\\nLYNCH,KEVIN (40x)\\nPENN,ALAN (35x)\\nHANSON,JULIENNE (32x)\\nRATTI,CARLO (31x)\\nSTEADMAN,PA (29x)',\n",
       " 'ALEXANDER,CH (62x)\\nALEXANDER,EH (59x)\\nHEALEY,P (59x)\\nFALUDI,ANDREAS (49x)\\nFORESTER,J (44x)\\nINNES,JD (41x)\\nFRIEND,J (34x)\\nFRIEDMANN,J (29x)\\nBERKE,P (23x)\\nLINDBLOM,C (23x)',\n",
       " 'BATTY,MA (243x)\\nWHITE,R (75x)\\nCOUCLELIS,H (57x)\\nCLARKE,K (56x)\\nWU,FJ (55x)\\nKLOSTERMAN,R (49x)\\nLANDIS,JR (48x)\\nMANDELBROT,BENOIT (48x)\\nTORRENS,PAUL (44x)\\nWILSON,A (43x)',\n",
       " 'KOLAREVIC,BRANKO (113x)\\nMENGES,ACHIM (86x)\\nGRAMAZIO,F (69x)\\nOXMAN,R (59x)\\nHENSEL,MU (52x)\\nOXMAN,NERI (50x)\\nBURRY,MC (40x)\\nRIPPMANN,MATTHAIS (36x)\\nBLOCK,PHILIPPE (29x)\\nFALLACARA,G (28x)',\n",
       " 'SMITH,R (25x)\\nSMITH,S (24x)\\nSMITH,B (16x)\\nSMITH,G (16x)\\nSMITH,D (15x)\\nSHAH,J (12x)\\nSMITH,A (12x)\\nSMITH,T (12x)\\nSMITH,J (9x)\\nSMITH,LAURAJANE (9x)',\n",
       " 'LEE,J (51x)\\nLEE,S (47x)\\nBLACKER,T (33x)\\nLEE,K (29x)\\nLEE,C (26x)\\nOWEN,SH (26x)\\nLEE,Y (23x)\\nGEORGE,P (22x)\\nMITTAL,S (22x)\\nSHEWCHUK,JP (20x)',\n",
       " 'REQUICHA,A (140x)\\nHOFFMANN,CH (104x)\\nROSSIGNAC,JAREK (77x)\\nHOPPE,HUGUES (61x)\\nMANTYLA,MJ (53x)\\nAMENTA,N (47x)\\nDEY,T (44x)\\nLIGHT,RA (44x)\\nOWEN,J (43x)\\nTAUBIN,G (36x)',\n",
       " 'EASTMAN,C (239x)\\nGRUBER,TB (62x)\\nSACKS,RAFAEL (47x)\\nAZHAR,SALMAN (43x)\\nLEE,G (39x)\\nZHANG,SAN-YUAN (35x)\\nBEETZ,J (34x)\\nPAUWELS,PIETER (33x)\\nSUCCAR,B (32x)\\nLIEBICH,T (30x)',\n",
       " 'PIEGL,LES (236x)\\nFARIN,GERALD (229x)\\nSEDERBERG,THOMAS (151x)\\nHOSCHEK,J (123x)\\nFAROUKI,RAMT (111x)\\nDE,B (96x)\\nFAUX,I (77x)\\nBOEHM,W (75x)\\nELBER,GERSHON (68x)\\nTILLER,W (60x)',\n",
       " 'SCHON,D-A (259x)\\nCROSS,NIGEL (232x)\\nLAWSON,BRYAN (132x)\\nDORST,KEE (117x)\\nGOLDSCHMIDT,GABRIELA (116x)\\nGERO,JOHN (102x)\\nSIMON,HD (101x)\\nGOEL,VK (79x)\\nSUWA,M (76x)\\nERICSSON,K (65x)',\n",
       " 'CHOI,BI (131x)\\nLEE,S (125x)\\nCHEN,Y (118x)\\nWANG,WENPING (115x)\\nLEE,J (111x)\\nLIU,Y-T (108x)\\nWANG,JG (107x)\\nWANG,XP (103x)\\nWANG,YT (101x)\\nCHEN,JCH (99x)',\n",
       " 'WOODBURY,R (53x)\\nSIMON,H (28x)\\nDAVIS,DB (21x)\\nEASTMAN,C (19x)\\nAKIN,O (17x)\\nAISH,ROBERT (16x)\\nDAVIS,MARGARET (14x)\\nOXMAN,R (13x)\\nSTINY,G (11x)\\nJOHNSON,S (10x)',\n",
       " 'MITCHELL,WB (30x)\\nKEENEY,RALPH (28x)\\nFOLEY,JA (26x)\\nSAATY,THOMAS (26x)\\nSIMON,HD (26x)\\nMILLER,G (25x)\\nCHURCH,R (24x)\\nSUTHERLAND,IVAN (20x)\\nWILSON,AD (20x)\\nHOPKINS,L (19x)',\n",
       " 'JENCKS,CA (25x)\\nROSSI,ALDO (22x)\\nTAFURI,M (22x)\\nHARVEY,DAVID (18x)\\nALBERTI,L (16x)\\nROWE,COLIN (15x)\\nFRAMPTON,KENNETH (14x)\\nBANHAM,REYNER (12x)\\nEISENMAN,PETER (12x)\\nFOX,M (12x)',\n",
       " 'GERO,JOHN (124x)\\nPAHL,GERHARD (78x)\\nGOEL,ASHOK (73x)\\nBROWN,D (69x)\\nMAHER,M (68x)\\nCHANDRASEKARAN,B (57x)\\nUMEDA,Y (56x)\\nSTONE,RJ (54x)\\nCHAKRABARTI,AK (46x)\\nULRICH,K (45x)']"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "clusterlabelsb = []\n",
    "clusterlabelsb_textblock = []\n",
    "counter = 0\n",
    "cluster_authors  = []\n",
    "cluster_citations = []\n",
    "linelength = 110\n",
    "#fullscoreTopCitations\n",
    "for x in range(0,len(fullscoreA)):\n",
    "    articleCountText = str(articleCount[x]) + ' articles'\n",
    "    authors = \"\\n\".join( [ (w[0].split()[0].upper() + ',' + w[0].split()[1].upper()+' ('+str(w[1])+'x)') for w in fullscoreA[x]])\n",
    "    cluster_authors.append(authors)\n",
    "    clustertext = \"MOST CITED AUTHORS: \" + \"; \".join( [ (w[0].split()[0].upper() + ',' + w[0].split()[1][0].upper()+'. ('+str(w[1])+'x)') for w in fullscoreA[x] ] )\n",
    "    citations = \"\\n\".join([ (w[0].upper() + ' ('+str(w[1])+'x)') for w in fullscoreTopCitations[x]])\n",
    "    cluster_citations.append(citations)\n",
    "    topcitationstext = \"TOP CITATIONS: \" + \"; \".join([ (w[0].upper() + ' ('+str(w[1])+'x)') for w in fullscoreTopCitations[x]])+''\n",
    "    #clustertext = clustertext + \"Words: \" + textwrap.fill(\"; \".join([ (w[0]+'('+str(w[1])+')') for w in filteredwords[x] if w[1]>10]),linelength)\n",
    "    #clusterlabels[x] = clusterlabels[x]+ '\\n' + clustertext\n",
    "    clusterlabelsb_textblock.append(articleCountText + \"\\n\" + clustertext + \"\\n\" + topcitationstext)\n",
    "    clustertext = textwrap.fill( clustertext,linelength)\n",
    "    #clustertext = articleCountText + \"\\n\" + clustertext + \"\\n\" + textwrap.fill(topcitationstext,linelength) #.upper()\n",
    "    clustertext = clustertext + \"\\n\" + textwrap.fill(topcitationstext,linelength) #.upper()\n",
    "    clusterlabelsb.append(clustertext)\n",
    "    #clusterlabelsb.append(\"\")\n",
    "    print(\"cluster: \" + str(counter))\n",
    "    #print(textwrap.fill(clustertext,70))\n",
    "    print(clustertext)\n",
    "    print(\"--------------------\")\n",
    "    counter +=1\n",
    "#display(clusterlabelsb_textblock)\n",
    "display(cluster_authors)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 287,
   "metadata": {},
   "outputs": [],
   "source": [
    "with open('exports/cluster_words.txt', 'w') as f:\n",
    "    f.writelines(\"%s\\n\" % line for line in clusterlabels_textblock)\n",
    "\n",
    "with open('exports/cluster_cited.txt', 'w') as f:\n",
    "    f.writelines(\"%s\\n\" % line for line in clusterlabelsb_textblock)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Plot the map\n",
    "\n",
    "Finally we have all the information to plot the map of articles with the clusters and their labels. For plotting we use `ggplot2` from `R` as before. It gives us a raw-looking graphic but we can export as PDF and edit it further with Affinity Designer, Adobe Illustrator or Inkscape."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 286,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Error in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y,  : \n",
      "  invalid font type\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/antonsavov/opt/anaconda3/envs/research_map_3/lib/python3.7/site-packages/rpy2/rinterface/__init__.py:145: RRuntimeWarning: There were 50 or more warnings (use warnings() to see the first 50)\n",
      "  warnings.warn(x, RRuntimeWarning)\n"
     ]
    }
   ],
   "source": [
    "%%R -i embedding,dfclust,my_cluster_colours_R,clusterlabels,clusterlabelsb,journalListLabel -w 594 -h 420 --units mm -r 300 --bg #F5F5F2\n",
    "#--bg #d3d3d3\n",
    "#-o labelpol,cltest --bg #231f20\n",
    "#-h 1600 -w 1600 -r 140 --bg #fbf8f1\n",
    "\n",
    "# Some imports:\n",
    "library(hrbrthemes)\n",
    "library(ggplot2)\n",
    "library(fields)\n",
    "#library(ggrepel) #was off in original from Noichl, used for geom_label_repel\n",
    "library(ggforce)\n",
    "#install.packages('ggalt')\n",
    "library(ggalt)\n",
    "library(stringr)\n",
    "#install.packages('clue')\n",
    "#library(clue)\n",
    "\n",
    "\n",
    "\n",
    "options(warn=0)# 0 zum anschalten\n",
    "\n",
    "#Get the cluster means:\n",
    "means <- aggregate(embedding[,c(\"x\",\"y\")], list(dfclust$cluster), mean)\n",
    "means <- data.frame(means)\n",
    "\n",
    "#And Variance, for the labels:\n",
    "test <- aggregate(embedding[,c(\"x\")], list(dfclust$cluster), var)\n",
    "test <- test[-1,]\n",
    "\n",
    "n=nrow(means)\n",
    "means <- means[-1,]\n",
    "\n",
    "#Make the colors: \n",
    "#mycolors <- c(\"#6abc70\", \"#0abcce\", \"#f9f4b8\", \"#ffe800\", \"#fbab19\", \"#f68121\", \"#ed1c24\", \"#ca205f\") \n",
    "#mycolors <- c(\"#4eeadf\",\"#e84a31\",\"#cc2ee7\",\"#bbe953\",\"#50aeec\")\n",
    "#saturated from ColorBrewer\n",
    "#mycolors <- c(\"#a6cee3\",\"#1f78b4\", \"#b2df8a\", \"#33a02c\", \"#fb9a99\", \"#e31a1c\", \"#fdbf6f\",\"#ff7f00\", \"#cab2d6\",\"#6a3d9a\", \"#ffff99\",\"#b15928\")\n",
    "\n",
    "#pastels from ColorBrewer\n",
    "#mycolors <- c(\"#8dd3c7\", \"#ffffb3\", \"#bebada\", \"#fb8072\", \"#80b1d3\", \"#fdb462\", \"#b3de69\", \"#fccde5\", \"#d9d9d9\", \"#bc80bd\", \"#ccebc5\", \"#ffed6f\")\n",
    "\n",
    "#pal <- colorRampPalette(sample(mycolors))\n",
    "s <- n-1\n",
    "#myGray <- c('#858583') # c('#95a5a6') # for the points outside of a cluster\n",
    "\n",
    "myGray <- c('#16457D') # ink blue\n",
    "#myNewColors <- sample(pal(s))\n",
    "myNewColors <- my_cluster_colours_R\n",
    "myPal <- append(myGray,myNewColors)\n",
    "\n",
    "labels <- clusterlabels            \n",
    "#labelsrows <- clusterlabelsrows\n",
    "labelsb <- clusterlabelsb  \n",
    "\n",
    "#circular markers:  \n",
    "library(gridExtra)\n",
    "\n",
    "circle <- polygon_regular(100)\n",
    "pointy_points <- function(x, y, size){\n",
    "  do.call(rbind, mapply(function(x,y,size,id) \n",
    "    data.frame(x=size*circle[,1]+x, y=size*circle[,2]+y, id=id),\n",
    "         x=x,y=y, size=size, id=seq_along(x), SIMPLIFY=FALSE))\n",
    "}\n",
    "\n",
    "#get density, to avoid overplotting\n",
    "embedding$density <- 1/ as.numeric(fields::interp.surface(MASS::kde2d(embedding$x, embedding$y), embedding[,c(\"x\",\"y\")]))\n",
    "\n",
    "# get for every label, whether it is in the + or - part of the x-axis:                        \n",
    "xpol <-abs(means[,c(\"x\")])/means[,c(\"x\")]\n",
    "ypol <-abs(means[,c(\"y\")])/means[,c(\"y\")]\n",
    "\n",
    "polfact <- 1.5\n",
    " \n",
    "#define label coordinates \n",
    "# build a circle for the labels:\n",
    "r <- 14.5\n",
    "sequence <- seq(from = 1, to = s, by = 1)\n",
    "angles <- 360/s*sequence\n",
    "angle <-(angles*(pi/180))\n",
    "         \n",
    "xlabl <- cos(angle)*r\n",
    "ylabl <- sin(angle)*r\n",
    "circlecord <- cbind(xlabl,ylabl)\n",
    "\n",
    "labelpol <-abs(circlecord[,c(\"xlabl\")])/circlecord[,c(\"xlabl\")]\n",
    "\n",
    "# install.packages(\"stringr\")\n",
    "# library(stringr)\n",
    "\n",
    "#make the label coordinates on two lines left and right of the plot\n",
    "halfclust = (s+1)%/%2\n",
    "col1 <- data.frame(xlabl=rep(-15.5,halfclust),ylabl=seq(-13,13,length=halfclust))\n",
    "col2 <- data.frame(xlabl=rep(15.5,s-halfclust),ylabl=seq(-13,13,length=s-halfclust))\n",
    "linecoords <- rbind(col1,col2)\n",
    "\n",
    "#make the label coordinates on one line on the right side of the plot\n",
    "linecoords <- data.frame(xlabl=rep(20.5,s),ylabl=seq(-19.5,19.5,length=s))\n",
    "\n",
    "circlecord <- linecoords\n",
    "\n",
    "coord_x=5\n",
    "coord_y=5\n",
    "                        \n",
    "# define circular markers:\n",
    "circular_annotations <- pointy_points(means$x, means$y, size=test$x*0.25+1)\n",
    "#\n",
    "embedding <- cbind(embedding,dfclust)\n",
    "filtered <- as.data.frame(subset(embedding, cluster >= 0))\n",
    "cltest <- filtered\n",
    "\n",
    "#Let's plot!\n",
    "mapdotshape = 16 # originally this was 20 (circle with fill and border). with 3(cross symbol) it takes a lot of time to render\n",
    "#this plots the points that belong to a cluster\n",
    "p <- ggplot(data=filtered, aes(x=x, y=y, color= as.factor(cluster),alpha=0.5))+ #, alpha='density'))+\n",
    "#pch=20 is bullet shape - https://stat.ethz.ch/R-manual/R-patched/library/graphics/html/points.html    \n",
    "geom_point(pch=mapdotshape,cex=1.5)+#, alpha = 0/density)+ \n",
    "#scale_color_manual(values = myPal)+ #this applies our colours to the cluster points\n",
    "scale_x_continuous(limits=c(-25,25),labels=NULL, breaks=seq(-25,25,length=16))+\n",
    "scale_y_continuous(limits=c(-15,15),labels=NULL, breaks=seq(-15,15,length=10))+\n",
    "#the lines below were disabled\n",
    "#geom_polygon(data=circular_annotations, aes(x,y,group=factor(id), fill = factor(id)),alpha=0.15)+\n",
    "#scale_fill_manual(values = myNewColors) #+      \n",
    "#this hides the legends                        \n",
    "#guides(alpha=FALSE, color=FALSE, fill=FALSE) #+\n",
    "#\n",
    "#this creates the coloured spots over the clusters that encircle them - takes time\n",
    "#stat_density2d(n=800,h=c(1.1,1.1))\n",
    "#\n",
    "xlim(-15, 50) + \n",
    "ylim(-20, 20) + \n",
    "NULL\n",
    "#end of the points plotting\n",
    "\n",
    "\n",
    "\n",
    "#ggplot_build prepares the plot for rendering - https://ggplot2.tidyverse.org/reference/ggplot_build.html\n",
    "#and also adds to p the coloured spots over the clusters by creating density contours around them\n",
    "q <- ggplot_build(p+stat_density2d(n=400,h=c(1.1,1.1)))$data[[2]]\n",
    "#looks like this filters out all points except those belonging to group 001 which is the first, outmost density contour \n",
    "#there is a problem because if first contour is comprised of several closed curves only one of them ends up being shown on the plot\n",
    "q <- q[str_detect(q$group, \"001\") == TRUE, ]\n",
    "# z <- max(test$x)\n",
    "# print(z)\n",
    "\n",
    "#this finds the point with minimum x for each cluster in the plot render data and gets its data in z\n",
    "o <- aggregate(q$x, list(q$group) , min)\n",
    "z <- subset(q,subset = q$x %in%  c(o$x))\n",
    "#this finds the point with maximum x for each cluster in the plot render data and gets its data in zmax\n",
    "omax <- aggregate(q$x, list(q$group) , max)\n",
    "zmax <- subset(q,subset = q$x %in%  c(omax$x))\n",
    "\n",
    "#this creates a dataframe with the points with min or max depending on which side of the origin each cluster lies\n",
    "c <- data.frame()\n",
    "count <- 1\n",
    "for (val in xpol) {\n",
    "if(val <0 ){\n",
    "    c <- rbind(c, z[count,])\n",
    "} else {\n",
    "    c <- rbind(c, zmax[count,])\n",
    "}\n",
    "    count <- count + 1\n",
    "}\n",
    "\n",
    "\n",
    "#these are the x and y coords of those points (leftmost or right most per cluster)\n",
    "#contactpoints <-  data.frame(c$x,c$y)\n",
    "contactpoints <-  data.frame(zmax$x,zmax$y)\n",
    "\n",
    "#Append every label to its best fit on that circle, using the hungarian algorithm:\n",
    "require(clue)\n",
    "#this gives all possible distances between the two sets of points, those on the circle and those in the cluster\n",
    "distances <- rdist(circlecord,contactpoints)\n",
    "#this give the indices of the points on the circle that fit closest to each cluster point\n",
    "sol <- solve_LSAP(t(distances))\n",
    "#this builds dataframe with the coordinates of each set of two matching points (cluster to circle)\n",
    "solo <- data.frame(cbind(mx=(contactpoints[,1]), my=(contactpoints[,2]), cx=(circlecord[sol, 1]), cy=(circlecord[sol, 2])))\n",
    "                        \n",
    "xcpol <-abs(solo$cx)/solo$cx\n",
    "\n",
    "#imprts the roboto condensed font for use in theme_ft_rc later\n",
    "#import_roboto_condensed()\n",
    "#import_titillium_web()\n",
    "\n",
    "#this block produces the final graph using the one produced by p as the base\n",
    "r <- \n",
    "p +\n",
    "#this adds the points outside of a cluster\n",
    "geom_point(data=subset(embedding, cluster == -1), aes(x=x, y=y),pch=mapdotshape,cex=1.5,alpha=0.5, color=myGray)+                      \n",
    "#this draws the polygons blobby coloured spots around each cluster. it uses the precalculated data in q\n",
    "#geom_polygon(data=q, aes(x,y ,group = as.factor(q$group),fill = as.factor(q$group)),alpha=0.3,linetype=1,size=0.6)+ # ,linetype=3, color=\"black\"\n",
    "#geom_polygon(data=q, aes(x,y ,group = as.factor(cluster),fill = as.factor(cluster)),alpha=0.3,linetype=1,size=0.6)+ # ,linetype=3, color=\"black\"\n",
    "#this draws the iso contours around the clusters in colour\n",
    "stat_density2d(aes(colour=as.factor(cluster)),n=400,h=c(1.1,1.1),linetype=1,size=0.6,alpha=0.3)+#,color=\"black\")+\n",
    "#this draws the iso contours around the clusters in black\n",
    "#stat_density2d(aes(colour=\"black\"),n=400,h=c(1.1,1.1),linetype=1,size=0.6,alpha=0.3)+#,color=\"black\")+\n",
    "##scale_fill_manual(values = myPal)+\n",
    "#this is a built in theme in hrbrthemes - https://www.rdocumentation.org/packages/hrbrthemes/versions/0.6.0/topics/theme_ipsum\n",
    "theme_ipsum()+\n",
    "#theme_ipsum_tw()+\n",
    "#theme_modern_rc()+\n",
    "## scale_fill_manual(values = myPal)\n",
    "#this hides the legends\n",
    "guides(alpha=FALSE, color=FALSE, fill=FALSE)+\n",
    "#\n",
    "labs(x=NULL, y=NULL, #x=\"UMAP-x\", y=\"UMAP-y\",\n",
    "    title=\"DIGITAL DESIGN IN ARCHITECTURE\",\n",
    "    subtitle=\"RESEARCH LANDSCAPE MAP 1975-2019\\nwith data from ~12,000 articles\\nmapped by who cited who\",\n",
    "    caption=paste(\"BY ANTON SAVOV\"))+   \n",
    "#set parameters for the plot title\n",
    "theme(plot.title = element_text(size=20, hjust = 0.5,colour = myGray))+ #, family=\"SC\", face=\"plain\"\n",
    "#set parameters for the plot subtitle\n",
    "theme(plot.subtitle = element_text(size=13, hjust = 0.5,colour = myGray))+ #, family=\"SC\", face=\"plain\"\n",
    "#this modifies the grid lines\n",
    "#theme(panel.grid.major = element_line(colour = \"grey\", linetype=\"dotted\", size=0.55),panel.grid.minor = element_blank())+\n",
    "theme(panel.grid.major = element_line(colour = myGray, linetype=\"dotted\", size=0.55),panel.grid.minor = element_blank())+\n",
    "#this sets the graph background to our background colour\n",
    "theme(plot.background = element_rect(fill = \"#F5F5F2\"))+\n",
    "#use expand_limits to create some space for more text\n",
    "#https://www.rdocumentation.org/packages/ggplot2/versions/3.1.0/topics/expand_limits\n",
    "#doesnt seem to affect the plot because in p we used scale_x_continuous and scale_y_continuous to set limits\n",
    "#expand_limits(x = c(-30,30),y = c(-30,30))+\n",
    "#expand_limits(x = c(r+10,0-r-10),y = c(r+10,0-r-10))+\n",
    "\n",
    "#fixes x scale to y scale in 1:1 by default - https://www.rdocumentation.org/packages/ggplot2/versions/3.1.0/topics/coord_fixed\n",
    "coord_fixed()+\n",
    "#creates an annotation - text or geometry - https://www.rdocumentation.org/packages/ggplot2/versions/3.1.0/topics/annotate\n",
    "#this one in particular does the little horizontal line extending from each cluster towards it label\n",
    "#annotate(\"segment\", x = c$x, y = c$y, xend = c$x+xcpol*0.5, yend =c$y, color= myNewColors,alpha=0.3, size = 1)+\n",
    "annotate(\"segment\", x = zmax$x, y = zmax$y, xend = zmax$x+0.5, yend =zmax$y, color= myNewColors,alpha=0.3, size = 1)+\n",
    "#and this one does the diagonal connecting lines starting at the end of the short one from above and ending in the label dots\n",
    "#annotate(\"segment\", x = c$x+xcpol*0.5, y =c$y, xend = solo$cx, yend = solo$cy, color=myNewColors,alpha=0.3, size = 1)+\n",
    "annotate(\"segment\", x = zmax$x+0.5, y =zmax$y, xend = solo$cx, yend = solo$cy, color=myNewColors,alpha=0.3, size = 1)+\n",
    "#this plots the dots where the label text will be, pch=16 is circle, pch=22 is square\n",
    "geom_point(data=solo, aes(x=cx, y=cy), color= myNewColors, alpha = 1,pch=16,size=7, stroke = 1)+\n",
    "#this labels the dots with the cluster number\n",
    "#annotate(\"text\", x = solo$cx, y = solo$cy, label = means[,c(\"Group.1\")], color=\"white\", fontface=\"bold\",  size=3.2, parse = TRUE, hjust=0.5)+\n",
    "\n",
    "annotate(\"text\", x = solo$cx, y = solo$cy, label = means[,c(\"Group.1\")], color=\"black\", fontface=\"bold\",  size=3.2, parse = TRUE, hjust=0.5)+\n",
    "## theme(plot.margin=grid::unit(c(0,0,0,0), \"mm\"))+\n",
    "#some left over attempt at the diagonal lines\n",
    "##annotate(\"segment\", x = (solo$mx+(test$x*0.25+1)*xcpol)+xcpol*0.3, y = solo$my, xend = solo$cx, yend = solo$cy, color= myNewColors, alpha = 0.25, size = 0.7)+        \n",
    "#this adds the labels, still dont know what parse means - +0.19\n",
    "#annotate(\"text\", x = solo$cx+xcpol*0.3, y = solo$cy, parse = FALSE, label = labels, color=myGray, fontface=\"bold\",  size=2, hjust=abs((xcpol+1)/2-1),vjust=0)+\n",
    "annotate(\"text\", x = solo$cx+0.7, y = solo$cy, parse = FALSE, label = labels, color=myGray, fontface=\"plain\",  size=2, hjust=0,vjust=0.8)+\n",
    "#this adds the sublabels - +0.3-0.5\n",
    "#annotate(\"text\", x = solo$cx+xcpol*0.3, y = solo$cy-0.1, parse = FALSE, label = labelsb, color=myGray, fontface=\"italic\",  size=2, hjust=abs((xcpol+1)/2-1),vjust=1)+            \n",
    "annotate(\"text\", x = solo$cx+17.2, y = solo$cy, parse = FALSE, label = labelsb, color=myGray, fontface=\"plain\",  size=2, hjust=0,vjust=0.8)+\n",
    "#could get this repel thing to work\n",
    "#geom_label_repel(data = c, aes(label = labels,\n",
    "#                    fill = myNewColors), color = 'white',\n",
    "#                    size = 1)\n",
    "#marks the cluster attachment points for the label leaders with black dots\n",
    "#geom_point(data=c, aes(x,y),color=\"black\")+\n",
    "#this sets our colours to the cluster dots\n",
    "scale_color_manual(values = myNewColors)+\n",
    "#this sets our colours to the cluster blobs\n",
    "scale_fill_manual(values = myNewColors)+\n",
    "#this sets the background to transparent\n",
    "#theme(plot.background=element_rect(fill=NA, colour=NA))+\n",
    "#this adds the label with the journals list - for vjust ref - https://stackoverflow.com/questions/7263849/what-do-hjust-and-vjust-do-when-making-a-plot-using-ggplot\n",
    "#x = -14.5, y = -14\n",
    "annotate(\"text\", x = -25, y = 8, parse = FALSE, label = journalListLabel, color=myGray, fontface=\"plain\",  size=2, hjust=0,vjust=0)+\n",
    "\n",
    "\n",
    "\n",
    "NULL\n",
    "#end of the plotting function of the cluster map\n",
    "\n",
    "\n",
    "#library(gridExtra)\n",
    "#grid.arrange(p,t, ncol = 1,  heights = c(1, 1))\n",
    "\n",
    "\n",
    "ggsave('exports/map-raw.png', plot = r,width=594,height=420, units = \"mm\",dpi=300, bg = \"#F5F5F2\")\n",
    "\n",
    "\n",
    "#pdf(\"exports/ClusteringUMap.pdf\",width=594/25.4,height=420/25.4) #,encoding='ISOLatin2.enc')#, paper='a4')# width = 12, height = 12) # Open a new pdf file\n",
    "#print(r)\n",
    "#dev.off()\n",
    "\n",
    "## my output commands to understand what is what\n",
    "r\n",
    "#print(n)\n",
    "#class(means)  # get class\n",
    "#sapply(means, class)  # get class of all columns\n",
    "#str(means)  # structure\n",
    "#summary(means)  # summary of dataframe\n",
    "#print(str(distances)) # \n",
    "#print(str(sol)) # \n",
    "#print(str(solo)) # \n",
    "#print(str(xcpol)) # \n",
    "#print(str(c)) # \n",
    "#print(head(test))\n",
    "#fix(sol)  # view spreadsheet like grid\n",
    "#rownames(means)  # row names\n",
    "#colnames(means)  # columns names\n",
    "#nrow(means)  # number of rows\n",
    "#ncol(means)  # number of columns\n",
    "#ggsave('plot-test.png', plot = last_plot(),width=594,height=420, units = \"mm\",dpi=300)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here is the image that is direclty output from the script above.\n",
    "![Raw Map of Digital Design in Architecture](images/map-raw.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# RESULT\n",
    "And here is again the image from the very top. The final version after the improvements done with Affinity Designer. Note that I used the algortihmically derived info for each cluster to give it a title. The titles are my interpretation and you can agree or disagree with them. In any case I would love to hear from you, email me at antonsavov@gmail.com or open an issue here.\n",
    "\n",
    "![Map of Digital Design in Architecture](images/map-digital-design-in-architecture.png)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}