Group170.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# COGS 108: Presidential Inaugural Address and 2016 Debate Analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# The Team\n",
    "Names and PIDs\n",
    "\n",
    "- **Kimiko Okumura**, A13577492\n",
    "- **Isabella Froman**, A13152119\n",
    "- **Jeff Xu**, A15653719\n",
    "- **Zhongrui Cao**, A15541281"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Introduction and Background\n",
    "\n",
    "As the United States and global political landscape grow more complicated and more polarized, we ask the question: how did we get here? Why are we here? Whether you are satified or discouraged by current affairs, understanding the current situation -- as well as any historical precendence -- may be key to moving forward toward a world you'd rather see. \n",
    "\n",
    "In politics, data science is essential to predicting election results and measuring public attitudes. It can be used to help the public form expectations on which policies to come -- and can even be used to characterize the political attitudes of a speech or region.\n",
    "\n",
    "We used 6 metrics (defined below) to understand, analyze, and compare the two types of speeches. For the debate data, we structured it as \"Trump vs. Clinton\" by dividing the corpus of three debate transcripts into datasets that define who is speaking. For the inaugural addresses, we used clustering to define three distinct eras in American history -- to be used to compare against other eras and to compare against the speech tactics of the 2016 presidential debates.\n",
    "\n",
    "\n",
    "### Research Question\n",
    "Our initial topic was on media engagement of newsworthy events -- exploring how long events of different types last in the public awareness. However, over the course of discussion and preparing for this topic, we found it very difficult to access data for Twitter. During this time, we began leaning toward text analysis and language processing, which prompted a brainstorm session for a new topic. Individual interest combined with suggestions from class (about analyzing song lyrics) inspired us to shift our focus to the political sphere -- and specifically how speech techniques have changed over time and how they differ across political party lines.\n",
    "\n",
    "**How do Presidential inaugural addresses change over time, and how does this compare to Clinton-Trump debates in the 2016 Presidential election?**\n",
    "\n",
    "To explore this, we are further asking: how might the speech patterns of the recent candidates (as well as presidents over time) reflect changing times and differing opinions? Does this have implications about other cultural changes and what the future of the Presidential Office might bring?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Metrics of Focus\n",
    "\n",
    "We will be examining 6 metrics to help us extract emerging patterns from inaugural addresses and debates. These 6 metrics are:\n",
    "\n",
    "   **1) Total Words** - How many words are said in total? The total length may represent cultural values about speech techniques such as detail vs. brevity.\n",
    "   \n",
    "   **2) Word Frequency** - Which words are used most frequently? This represents the focus of each candidate, and can be used to loosely characterize the candidate or president's personality -- or in the case of clustering, define the cultural values of an era.\n",
    "   \n",
    "   **3) Average Word Length** - How long (and by extension, how complex) are the words each candidate is using? This represents simplicity, and may have implications about the candidate's intelligence or the intended audience of the speech.\n",
    "\n",
    "   **4) Narcissism Index** - What is the proportion of occurrences of “I,” “me,” and other self-referential words compared to other words? This represents a personality trait believed to be in many successful people -- a lot of discussion of Donald Trump explores topics such as narcissism.\n",
    "\n",
    "   **5) Lexical Density / Vocabulary Distribution** - What is the distribution of different kinds of words used? This represents the variety of words the candidate uses -- which may reflect the candidate's perceived intelligence.\n",
    "\n",
    "   **6) Sentiment Analysis** - What is the emotional content of the speech? This represents the emotional values of each candidate, and also loosely represents the candidate's personality."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Hypothesis\n",
    "\n",
    "We think that Donald Trump's debates will have a lower average word length, higher degrees of narcissism, higher density of (fewer unique) words, extreme emotional sentiment (positive and negative). We believe this will be due to Trump's lack of political experience, as well as his reputation as a brash and decisive person.\n",
    "\n",
    "In contrast, Hillary Clinton's debates will have a higher average word length and sentence length, lower degrees of narcissism, lower density of (more unique) words, and more neutral in emotional sentiment. We believe this will be due to Clinton's experience as a politician, as well as her reputation as an educated and calculating person.\n",
    "\n",
    "In our time-scale analysis of inaugural addresses, we hypothesize that word frequency increases, average word length and sentence length decrease, narcissism increases, and lexical density increases over time. We believe this due to an increasing appeal to the public (including those not educated about political affairs).\n",
    "\n",
    "For each of these comparative analyses, the null hypothesis is that the debate techniques (defined by our 6 metrics) of debates are not significantly different from inaugural addresses. Our alternative hypothesis is that there is a statistically significant difference between the debate techniques and the inaugural address techniques."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Prior work and research on text and sentiment analysis:\n",
    "\n",
    "\n",
    "References (with inline links):\n",
    "\n",
    "1) [Natural Language Toolkit (NLTK)](https://www.nltk.org/)\n",
    "This package has been instrumental to our analysis in providing a framework for analysis speech and sentiment.\n",
    "\n",
    "\n",
    "2) [Trump-specific transcript data and analysis](https://factba.se/)\n",
    "This website conducts data analysis on Trump's speeches, as well as collects/analyzes data on other conservative politicians. This set some examples for what our project could look like.\n",
    "\n",
    "\n",
    "3) [NLP on Inaugural Address](https://towardsdatascience.com/intro-to-nlp-using-inaugural-speeches-of-presidents-8c7ca32cbdfe)\n",
    "This is a guide to using natural language processing -- using inaugural addresses as the corpus. This was VERY helpful in our preliminary research.\n",
    "\n",
    "\n",
    "4) [Example of Inaugural Speech Analysis](https://pradeepadhokshaja.wordpress.com/2017/05/01/analysis-of-inaugural-speech-data-part-1/)\n",
    "This is an example of an analysis of inaugural addresses -- it also set the tone for our project.\n",
    "\n",
    "\n",
    "5) [Sentiment Analysis on State of the Union Speeches](https://towardsdatascience.com/sentiment-of-the-union-analyzing-presidential-state-of-the-union-addresses-with-python-2a8667a578b9)\n",
    "This is an example of sentiment analysis of State of the Union Presidential Addresses -- another example that inspired our project.\n",
    "\n",
    "6) [Analysis and Comparison of Trump's Inaugural Speech](https://towardsdatascience.com/a-data-science-exercise-how-similar-is-trump-to-other-presidents-based-on-inaugural-speeches-8fa7f534a5fb)\n",
    "A further example of inaugural address analysis -- this time involving Donald Trump specifically.\n",
    "\n",
    "\n",
    "7) [Measuring candidate narcissism](https://www.datascience.com/blog/debates/)\n",
    "A guide to measuring narcissism in a speech corpus -- particularly focused on presidential speeches. Crucial to our project and goals."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Dataset(s)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Clinton-Trump Debate Datasets\n",
    "\n",
    "**\"September 26, 2016: The First Clinton-Trump Presidential Debate\"**\n",
    "- Link to the dataset: https://www.debates.org/voter-education/debate-transcripts/september-26-2016-debate-transcript/\n",
    "- Number of observations: Textual analysis\n",
    "\n",
    "This is a transcript of the *first* Presidential debate between Hillary Clinton and Donald Trump in the 2016 United States Presidential Election. The moderator is Lester Holt from NBC News, who prompts the candidates with questions. These debates cover several topics that are key to defining candidates' policies, personalities, and goals -- with the intention of demonstrating competence and persuading undecided voters.\n",
    "\n",
    "**October 9, 2016: The Second Clinton-Trump Presidential Debate**\n",
    "- Link to the dataset: https://www.debates.org/voter-education/debate-transcripts/october-9-2016-debate-transcript/\n",
    "- Number of observations: Textual analysis\n",
    "\n",
    "This is a transcript of the *second* Presidential debate between Hillary Clinton and Donald Trump in the 2016 United States Presidential Election. The moderators are Anderson Cooper from CNN and Martha Raddatz from ABC News, who prompts the candidates with questions. These debates cover several topics that are key to defining candidates' policies, personalities, and goals -- with the intention of demonstrating competence and persuading undecided voters.\n",
    "\n",
    "**October 19, 2016: The Third Clinton-Trump Presidential Debate**\n",
    "- Link to the dataset: https://www.debates.org/voter-education/debate-transcripts/october-19-2016-debate-transcript/\n",
    "- Number of observations: Textual analysis\n",
    "\n",
    "This is a transcript of the *third and final* Presidential debate between Hillary Clinton and Donald Trump in the 2016 United States Presidential Election. The moderator is Chris Wallace from Fox News, who prompts the candidates with questions. These debates cover several topics that are key to defining candidates' policies, personalities, and goals -- with the intention of demonstrating competence and persuading undecided voters.\n",
    "\n",
    "### Inaugural Addresses Dataset\n",
    "\n",
    "**Dataset Name: Presidential Inaugural Addresses**\n",
    "- Link to the dataset: https://www.kaggle.com/adhok93/presidentialaddress?fbclid=IwAR0UXh5kvnZXTMcIDO9t7DEIhn-L7QJMpVMLWTxF55jdhqu3EAdiJiPKVW4\n",
    "- Number of observations: 58 -- There are 58 separate speeches.\n",
    "\n",
    "This is a .csv file of the President Inaugural Addresses since the first President George Washington. The .csv file contains the name of each president, which (first or second) address it is, the date of the speech, and the full transcript of the inaugural address. The purpose of these speeches is to address the country after being voted into office -- usually to inspire confidence in the president's capabilities and set the tone for their upcoming term as President of the United States.\n",
    "\n",
    "\n",
    "### Combining Datasets\n",
    "The three debate transcripts will be compared with each other to find any important words, patterns, or differences among candidates, and then will be combined with each other to create a \"Clinton vs. Trump\" Corpus. The inaugural addresses will be kept distinct -- measuring each of our metrics -- and then compared to the debate corpus.\n",
    "\n",
    "However, inaugural addresses and presidential debates are fundamentally different speeches. Inaugural addresses are announcements, while debates are much more persuasive in nature. There are also much more rigid rules about debates than inaugural speeches. For this reason, the two types will not be compared on every single metric, and instead will be discussed in separate terms."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Setup"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To set up, we imported our necessary packages and modules. Some of these include standard python packages such as Pandas and Numpy, but other were more specific to our project, such as the Natural Language Processing Toolkit (NLTK) and data visualization toolkits such as Seaborn."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: mosestokenizer in /anaconda3/lib/python3.7/site-packages (1.0.0)\n",
      "Requirement already satisfied: openfile in /anaconda3/lib/python3.7/site-packages (from mosestokenizer) (0.0.7)\n",
      "Requirement already satisfied: docopt in /anaconda3/lib/python3.7/site-packages (from mosestokenizer) (0.6.2)\n",
      "Requirement already satisfied: toolwrapper in /anaconda3/lib/python3.7/site-packages (from mosestokenizer) (0.4.1)\n",
      "Requirement already satisfied: TextBlob in /anaconda3/lib/python3.7/site-packages (0.15.3)\n",
      "Requirement already satisfied: nltk>=3.1 in /anaconda3/lib/python3.7/site-packages (from TextBlob) (3.4)\n",
      "Requirement already satisfied: six in /anaconda3/lib/python3.7/site-packages (from nltk>=3.1->TextBlob) (1.12.0)\n",
      "Requirement already satisfied: singledispatch in /anaconda3/lib/python3.7/site-packages (from nltk>=3.1->TextBlob) (3.4.0.3)\n"
     ]
    }
   ],
   "source": [
    "#Install\n",
    "!pip install mosestokenizer\n",
    "!pip install TextBlob"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[nltk_data] Downloading package punkt to\n",
      "[nltk_data]     /Users/kimikookumura/nltk_data...\n",
      "[nltk_data]   Package punkt is already up-to-date!\n",
      "[nltk_data] Downloading package stopwords to\n",
      "[nltk_data]     /Users/kimikookumura/nltk_data...\n",
      "[nltk_data]   Package stopwords is already up-to-date!\n",
      "[nltk_data] Downloading package vader_lexicon to\n",
      "[nltk_data]     /Users/kimikookumura/nltk_data...\n",
      "[nltk_data]   Package vader_lexicon is already up-to-date!\n",
      "[nltk_data] Downloading package perluniprops to\n",
      "[nltk_data]     /Users/kimikookumura/nltk_data...\n",
      "[nltk_data]   Package perluniprops is already up-to-date!\n"
     ]
    }
   ],
   "source": [
    "# General packages\n",
    "import string\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import math\n",
    "import random\n",
    "\n",
    "# Language analysis: Import nltk package, PennTreeBank word \n",
    "# tokenizer, Frequency Distribution, English language stopwords\n",
    "import nltk\n",
    "from nltk.tokenize import RegexpTokenizer\n",
    "from nltk.tokenize import word_tokenize\n",
    "from nltk.tokenize import sent_tokenize\n",
    "from nltk.probability import FreqDist\n",
    "from nltk.corpus import stopwords\n",
    "from nltk.sentiment.vader import SentimentIntensityAnalyzer \n",
    "from textblob import TextBlob\n",
    "from nltk.stem.snowball import SnowballStemmer\n",
    "from mosestokenizer import MosesTokenizer, MosesDetokenizer\n",
    "\n",
    "\n",
    "nltk.download('punkt')\n",
    "nltk.download('stopwords')\n",
    "nltk.download('vader_lexicon')\n",
    "nltk.download('perluniprops')\n",
    "\n",
    "# General Analysis\n",
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "from datetime import datetime\n",
    "from statistics import mean\n",
    "from scipy import stats\n",
    "\n",
    "# Data Visualization\n",
    "import matplotlib\n",
    "import matplotlib.pyplot as plt\n",
    "from matplotlib import colors\n",
    "from matplotlib.ticker import PercentFormatter\n",
    "import seaborn as sns\n",
    "plt.rcParams.update({'font.size': 10})\n",
    "\n",
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.metrics import classification_report, \\\n",
    "                            precision_recall_fscore_support\n",
    "from datetime import datetime\n",
    "from nltk.stem.snowball import SnowballStemmer\n",
    "\n",
    "# Web Scraping\n",
    "import bs4\n",
    "from bs4 import BeautifulSoup\n",
    "import requests\n",
    "\n",
    "#Regular Expression\n",
    "import re\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: TextBlob in /anaconda3/lib/python3.7/site-packages (0.15.3)\r\n",
      "Requirement already satisfied: nltk>=3.1 in /anaconda3/lib/python3.7/site-packages (from TextBlob) (3.4)\r\n",
      "Requirement already satisfied: six in /anaconda3/lib/python3.7/site-packages (from nltk>=3.1->TextBlob) (1.12.0)\r\n",
      "Requirement already satisfied: singledispatch in /anaconda3/lib/python3.7/site-packages (from nltk>=3.1->TextBlob) (3.4.0.3)\r\n"
     ]
    }
   ],
   "source": [
    "!pip install TextBlob"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Set Up Inaugural Speeches Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Read in the large Inaugural speech corpus into a DataFrame\n",
    "\n",
    "inaug_dat = pd.read_csv(\"Data/inaug_speeches.csv\", encoding = \"utf-8\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Set Up Clinton vs. Trump Debate Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Read in the three debates datasets of focus\n",
    "\n",
    "d1 = list(open(\"Data/Debates from 2016/19-2016-debate.txt\", \"r\", \\\n",
    "               encoding='utf-8').readlines())\n",
    "d2 = list(open(\"Data/Debates from 2016/26-2016-debate.txt\", \"r\", \\\n",
    "               encoding='utf-8').readlines())\n",
    "d3 = list(open(\"Data/Debates from 2016/9-2016-debate.txt\", \"r\", \\\n",
    "               encoding='utf-8').readlines())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data Wrangling, Cleaning, and Pre-processing"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Data cleaning occured in several steps -- as we had two distinct types of datasets, the cleaning required for each was very different.\n",
    "\n",
    "For the inaugural data, several trancsript cells of the .csv had strange symbols or unicode strings, which needed to be removed. We wrote a function to remove it, and used other general data cleaning techniques such as .lower(). We also took out stopwords, defined tokens, and added term numbers to each speech (to specify which term each president was spekaing for).\n",
    "\n",
    "For the debate data, all three of the debate transcripts had a dialogue format -- similar to a script -- since each debate had 3 or 4 people speaking. This meant we had to take out all of \"CLINTON:\" and \"TRUMP:\" headers, as well as keep the speakers distinct in their own sets for comparative analysis. In addition, we removed stopwords and made the corpus lowercase."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## > Inaugural Data\n",
    "We set up functions to clean up and interpret our data. We realized that in our inaugural speeches, there were random unicode strings floating around along with random \"? ? ?\" strings so we created these to correspondingly filter them out. We also needed to convert the string dates into comparable dates, which was done using the datetime.strptime method."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define function cleanText\n",
    "# Remove alphanumeric strings\n",
    "def cleanText( inputText ):\n",
    "    return re.sub(r'[^\\w ]', '',inputText);\n",
    "\n",
    "# Define removeUnicode function\n",
    "# Does a global search, removes Unicode substrings\n",
    "\n",
    "def removeUnicode(str):\n",
    "    str = str.lower()\n",
    "    mod_str = re.sub(r'[<|>|+]','',str)\n",
    "    return re.sub(r'u+00+[0-9]+[0-9]','',mod_str);\n",
    "        \n",
    "# Define removeQ function\n",
    "# Takes a string, returns a modified string without question marks \n",
    "def removeQ(str):\n",
    "    return str.replace(\"?\",\"\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'fjdskljsdkfdsfdfdd hello jfkdljd '"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#removeUnicode() test\n",
    "test = \"fjd+skl+jsdk<fdsfd>fdd hello <u+0094>jfkdljd <u00938>\"\n",
    "removeUnicode(test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define English stopwords \n",
    "stopWords = set(stopwords.words('english'))\n",
    "\n",
    "# Define function tokenizeAdd\n",
    "# Pass in speech transcript, change everything to lowercase,\n",
    "# remove stopwords and return the rest of the word tokens\n",
    "def tokenizeAdd (raw):\n",
    "    result = []\n",
    "    for w in raw:\n",
    "        if w.lower() not in stopWords:\n",
    "            result.append(w)\n",
    "    return result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define tokenizer, modify it to remove punctuation.\n",
    "tokenizer = RegexpTokenizer(r'\\w+')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "#ONLY RUN ONCE\n",
    "\n",
    "# Loop through all inaugural address speeches\n",
    "# Remove unicode substrings\n",
    "# Remove punctuation\n",
    "\n",
    "inaug_word_tokens = []\n",
    "inaug_word_tokens_nostop = []\n",
    "inaug_sent = []\n",
    "\n",
    "# Create list for word tokens with stop words\n",
    "for i in inaug_dat.index:\n",
    "    inaug_dat.iloc[i,4] = removeUnicode(str(inaug_dat.iloc[i,4]))\n",
    "    inaug_dat.iloc[i,4] = removeQ(inaug_dat.iloc[i,4])\n",
    "    inaug_dat.iloc[i,4] = inaug_dat.iloc[i,4].replace(\",\",\"\")\n",
    "    inaug_dat.iloc[i,4] = inaug_dat.iloc[i,4].replace(\":\",\"\")\n",
    "    inaug_dat.iloc[i,4] = inaug_dat.iloc[i,4].replace(\"-\",\" \")\n",
    "    inaug_dat.iloc[i,4] = inaug_dat.iloc[i,4].replace(\";\",\"\")\n",
    "    inaug_dat.iloc[i,4] = inaug_dat.iloc[i,4].replace(\"(\",\"\")\n",
    "    inaug_dat.iloc[i,4] = inaug_dat.iloc[i,4].replace(\")\",\"\")\n",
    "    inaug_sent.append(inaug_dat.iloc[i,4])\n",
    "    inaug_dat.iloc[i,4] = inaug_dat.iloc[i,4].replace(\".\",\"\")\n",
    "    inaug_word_tokens.append(nltk.word_tokenize(inaug_dat.iloc[i,4]))  \n",
    "\n",
    "# Create list for word tokens without stop words\n",
    "for e in inaug_word_tokens:\n",
    "    for w in inaug_word_tokens[i]:\n",
    "        if w.lower() not in stopWords:\n",
    "            inaug_word_tokens_nostop.append(w)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Add in the term number as a column\n",
    "#ONLY RUN ONCE\n",
    "\n",
    "term = list(range(1, 59))\n",
    "inaug_dat['Term'] = term\n",
    "\n",
    "inaug_dat.columns = ['Trash','Name','Inaugural Address','Date',\\\n",
    "                     'Text','Term']\n",
    "\n",
    "inaug_dat = inaug_dat.drop(columns=['Trash'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## > Debate Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "currentSpeaker = \"\"\n",
    "whatTrumpSaid = \"\"\n",
    "whatClintonSaid = \"\"\n",
    "\n",
    "##reading what C and T said\n",
    "#19 debate\n",
    "for line in d1:\n",
    "    if \"TRUMP\" in line:\n",
    "        whatTrumpSaid = whatTrumpSaid + line\n",
    "        currentSpeaker = \"TRUMP\"\n",
    "        continue\n",
    "    if \"CLINTON\" in line:\n",
    "        whatClintonSaid = whatClintonSaid + line\n",
    "        currentSpeaker = \"CLINTON\"\n",
    "        continue\n",
    "    if \"WALLACE\" in line:\n",
    "        currentSpeaker = \"WALLACE\"\n",
    "        continue\n",
    "    if currentSpeaker == \"TRUMP\":\n",
    "        whatTrumpSaid = whatTrumpSaid + line\n",
    "    if currentSpeaker == \"CLINTON\":\n",
    "        whatClintonSaid = whatClintonSaid + line\n",
    "    if currentSpeaker == \"WALLACE\":\n",
    "        continue\n",
    "\n",
    "#26 debate\n",
    "for line in d2:\n",
    "    if \"TRUMP\" in line:\n",
    "        whatTrumpSaid = whatTrumpSaid + line\n",
    "        currentSpeaker = \"TRUMP\"\n",
    "        continue\n",
    "    if \"CLINTON\" in line:\n",
    "        whatClintonSaid = whatClintonSaid + line\n",
    "        currentSpeaker = \"CLINTON\"\n",
    "        continue\n",
    "    if \"HOLT\" in line:\n",
    "        currentSpeaker = \"HOLT\"\n",
    "        continue\n",
    "    if currentSpeaker == \"TRUMP\":\n",
    "        whatTrumpSaid = whatTrumpSaid + line\n",
    "    if currentSpeaker == \"CLINTON\":\n",
    "        whatClintonSaid = whatClintonSaid + line\n",
    "    if currentSpeaker == \"HOLT\":\n",
    "        continue\n",
    "\n",
    "#9 debate\n",
    "for line in d3:\n",
    "    if \"TRUMP\" in line:\n",
    "        whatTrumpSaid = whatTrumpSaid + line\n",
    "        currentSpeaker = \"TRUMP\"\n",
    "        continue\n",
    "    if \"CLINTON\" in line:\n",
    "        whatClintonSaid = whatClintonSaid + line\n",
    "        currentSpeaker = \"CLINTON\"\n",
    "        continue\n",
    "    if \"RADDATZ\" in line:\n",
    "        currentSpeaker = \"RADDATZ\"\n",
    "        continue\n",
    "    if \"COOPER\" in line:\n",
    "        currentSpeaker = \"COOPER\"\n",
    "        continue\n",
    "    if \"QUESTION\" in line:\n",
    "        currentSpeaker = \"QUESTION\"\n",
    "        continue\n",
    "    if currentSpeaker == \"TRUMP\":\n",
    "        whatTrumpSaid = whatTrumpSaid + line\n",
    "    if currentSpeaker == \"CLINTON\":\n",
    "        whatClintonSaid = whatClintonSaid + line\n",
    "    if currentSpeaker == \"RADDATZ\":\n",
    "        continue\n",
    "    if currentSpeaker == \"COOPER\":\n",
    "        continue\n",
    "    if currentSpeaker == \"QUESTION\":\n",
    "        continue"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Modify the tokenizer to remove punctuations\n",
    "tokenizer = RegexpTokenizer(r'\\w+')\n",
    "\n",
    "# Clean data to remove punctuation unecessary for analysis\n",
    "whatTrumpSaid = whatTrumpSaid.replace(\"TRUMP: \", \"\")\n",
    "whatTrumpSaid = whatTrumpSaid.replace(\"\\n\",\"\")\n",
    "whatTrumpSaid = whatTrumpSaid.replace(\"—\",\" \")\n",
    "whatTrumpSaid = whatTrumpSaid.replace(\":\",\" \")\n",
    "whatTrumpSaid = whatTrumpSaid.replace(\"[crosstalk]\",\" \")\n",
    "\n",
    "whatTrumpSaid = whatTrumpSaid.replace(\"[applause]\",\" \")\n",
    "whatTrumpSaid = whatTrumpSaid.lower()\n",
    "TrumpTokens = tokenizer.tokenize(whatTrumpSaid)\n",
    "\n",
    "whatClintonSaid = whatClintonSaid.replace(\"CLINTON: \", \"\")\n",
    "whatClintonSaid = whatClintonSaid.replace(\"\\n\",\"\")\n",
    "whatClintonSaid = whatClintonSaid.replace(\"–\",\" \")\n",
    "whatClintonSaid = whatClintonSaid.replace(\"[crosstalk]\",\"\")\n",
    "whatClintonSaid = whatClintonSaid.lower()\n",
    "ClintonTokens = tokenizer.tokenize(whatClintonSaid)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Remove stopwords and storing it into a new list so the\n",
    "# original is still accessible\n",
    "stopWords = set(stopwords.words('english'))\n",
    "TrumpNoStopTokens = []\n",
    "ClintonNoStopTokens = []\n",
    "\n",
    "for w in TrumpTokens:\n",
    "    if w.lower() not in stopWords:\n",
    "        TrumpNoStopTokens.append(w)\n",
    "        \n",
    "for w in ClintonTokens:\n",
    "    if w.lower() not in stopWords:\n",
    "        ClintonNoStopTokens.append(w)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data Analysis & Results"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We looked at the Inaugural Addresses and 2016 Clinton-Trump debates for all seven metrics previously outlined. For each relevant metric, we also compared the addresses and debates to see if the debate techiniques fit in with what could be considered \"Presidential Speech Tactics\" as defined by the average values of the Inaugural Addresses."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Clustering Inaugural Speeches\n",
    "In the following section, we plan to analyze presidential inaugural speeches and how \n",
    "they have changed over time in terms of relevant content. Specifically, we will accompish\n",
    "this through scikit learn's KMeans method combined with the TFIDF functions. Instead of clustering every single inaugural address, we decided that discrete intervals of time periods would be better to visualize the differences. We found that if we included every single inagural speech, the clusters would be very inconsistent upon the changing of very small parameters."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "stdbuf was not found; communication with perl may hang due to stdio buffering.\n"
     ]
    }
   ],
   "source": [
    "snowStemmer = SnowballStemmer(\"english\").stem\n",
    "detokenizer = MosesDetokenizer()\n",
    "\n",
    "# Remove non-alpha-numeric strings\n",
    "# Does not tokenize\n",
    "def cleanText( inputText ):\n",
    "    inputText = removeUnicode( inputText )\n",
    "    return re.sub(r'[^\\w\\- ]', '', inputText );\n",
    "\n",
    "# Convert string into date-time object\n",
    "def convertDate( timeStr ):\n",
    "    timeStr = timeStr.strip()\n",
    "    date = 0;\n",
    "    try:\n",
    "        date = datetime.strptime( timeStr, '%A, %B %d, %Y' )\n",
    "    except:\n",
    "        date = datetime.strptime( timeStr, '%B %d, %Y' )\n",
    "    return date\n",
    "\n",
    "# Will handle tokenization, stop words, and stemming\n",
    "def processText( text ):\n",
    "    tokenized = [ word for sent in nltk.sent_tokenize(text) for\\\n",
    "                 word in word_tokenize(sent) ]\n",
    "    stems = [ snowStemmer(t) for t in tokenized ]\n",
    "    stems = tokenized\n",
    "    return stems\n",
    "\n",
    "# Function to reconcatenate tokenized strings \n",
    "def detokenizeText( tokenized ):\n",
    "    return detokenizer.detokenize( tokenized, return_str = True )\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Next, we need to re-organize the data so that we can index speeches based on the presidents' names. Instead of using pandas here, we resort to a dictionary for the sake of storing datetime objects that offer us comparable dates - a feature that will allow us to sort out the dates and compare them. With these datetime objects then, we can group disecrete sets of time periods."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "# This will be a dictionary containing the speeches of specific presidents,\n",
    "# Where keys will map to a list of tuples in the format ( date, text )\n",
    "# Examples will be provided below\n",
    "\n",
    "inaug_database=dict()\n",
    "\n",
    "# Set up the database so that we can organize by president name\n",
    "# If a president name already exists, then add to that list\n",
    "# Otherwise, make a new list for that president\n",
    "for index, row in inaug_dat.iterrows():\n",
    "    name = row.Name\n",
    "    date = row.Date\n",
    "    text = row.Text\n",
    "    if name in inaug_database:\n",
    "        inaug_database[name].append( ( date, convertDate( date ),\\\n",
    "                                      text, cleanText( text ) ) )\n",
    "    else:\n",
    "        inaug_database[name] = [ ( date, convertDate( date ), text,\\\n",
    "                                  cleanText( text ) ) ]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now that this dictionary is set up we can specify certain date ranges to pick from. In this way, we can choose from the data so as to create discrete clusters based on time period. For the cell block below, we will be picking the year range from 1800-1830, 1900-1925, 1941-1950, and 1990 - 2015. The 1800-1830 period is used to pick the oldest speeches, the 1900-1925 speeches were picked because of the World War I era, and 1941-1950 for the World War II era. We want to see if it is possible to differniate between our speeches just based on how historical contexts affects word usage and see the relevant and \"weighty\" words surrounding those inaugural speeches.\n",
    "\n",
    "Clustering will also allow us to see how speeches from different eras may be different/similar from each other.\n",
    "\n",
    "We will be appending the speeches within these year ranges within our speeches[] list."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "56525\n"
     ]
    }
   ],
   "source": [
    "# The year ranges we are specifying    \n",
    "yearRanges = [ (1800, 1835), (1925, 1955), (1980, 2015) ]\n",
    "#yearRanges = [ (1700,2018)]\n",
    "\n",
    "tokenizedText = []\n",
    "presidents = []\n",
    "speeches = []\n",
    "years = []\n",
    "wordCount = 0\n",
    "\n",
    "# Find text within valid year ranges\n",
    "for i, t in enumerate( inaug_database ):\n",
    "    entry = inaug_database[ t ]\n",
    "    for i, speech in enumerate( entry ):\n",
    "        \n",
    "        valid = False\n",
    "        year = speech[ 1 ].year\n",
    "    \n",
    "        for r in yearRanges:\n",
    "            \n",
    "            # Check if the year is within a valid range\n",
    "            \n",
    "            if year >= r[ 0 ] and year <= r[ 1 ]:\n",
    "                valid = True\n",
    "                break\n",
    "                \n",
    "        if ( valid ):\n",
    "            processed = processText( speech[ 3 ] ) \n",
    "            presidents.append( t )\n",
    "            speeches.append( speech[ 3 ] )\n",
    "            years.append( year )\n",
    "\n",
    "            tokenizedText.append( processed )\n",
    "            wordCount += len( processed )\n",
    "\n",
    "# Checking how many words we have total\n",
    "print( wordCount )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now that we created a list of speeches within these year ranges, we can begin our analyses. In order to work with cluster, we will create a TFIDF matrix for the speeches. This allows us to create a structure that contains the list of our speeches while containing specific tfidf values for each \"term\". After creating the matrix, we run it thorugh scikitlearn's KMeans clustering method to discover the different clusters among the speeches.\n",
    "\n",
    "[This tutorial](http://brandonrose.org/clustering) allowed us to understand how the clustering worked, as we follow through similar steps."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/anaconda3/lib/python3.7/site-packages/sklearn/feature_extraction/text.py:300: UserWarning: Your stop_words may be inconsistent with your preprocessing. Tokenizing the stop words generated tokens [\"'d\", \"'ll\", \"'re\", \"'s\", \"'ve\", 'could', 'might', 'must', \"n't\", 'need', 'sha', 'wo', 'would'] not in stop_words.\n",
      "  'stop_words.' % sorted(inconsistent))\n"
     ]
    }
   ],
   "source": [
    "# Specify stop words for our vectorizer\n",
    "from nltk.corpus import stopwords\n",
    "from sklearn.cluster import KMeans\n",
    "\n",
    "num_clusters = len( yearRanges ) + 1 # The number of clusters\n",
    "                                     # we plan to make\n",
    "\n",
    "# Create a set of stopwords\n",
    "# We include mr, and, and mrs because these were showing up as\n",
    "# signficant words but shouldn't be\n",
    "stop_words = set( stopwords.words(\"english\") ) | \\\n",
    "                set([\"mr\",\"and\",\"mrs\"])\n",
    "\n",
    "# Initialize vectorizer for tfidf\n",
    "# Specifying max df to 80% and min to 20%\n",
    "vectorizer = TfidfVectorizer( sublinear_tf = True,\n",
    "                        analyzer = 'word',\n",
    "                        max_features = 35000,\n",
    "                        tokenizer = word_tokenize,\n",
    "                        stop_words = stop_words,\n",
    "                        max_df = 0.8, min_df = 0.2 )\n",
    "\n",
    "tfidf_matrix = vectorizer.fit_transform( speeches ).toarray()\n",
    "km = KMeans( n_clusters = num_clusters )\n",
    "clusters = km.fit( tfidf_matrix )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.externals import joblib\n",
    "\n",
    "joblib.dump(km, 'doc_cluster.pkl')\n",
    "km = joblib.load('doc_cluster.pkl')\n",
    "clusters = km.labels_.tolist()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The block above compiles our cluster results based off the speeches and loads them so that we may analyze them to the specific word content.\n",
    "\n",
    "If we want to see the results of the clustering, lets create a dataframe and includes the paramaeters of the speech year, president, and cluster label:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "speech_data = {'year':years, 'president': presidents, \\\n",
    "               'cluster': clusters}\n",
    "cluster_frame = pd.DataFrame( speech_data, columns = \\\n",
    "                             ['year','president','cluster'] )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The cluster results show that there is a difference in the way speeches are written within different eras! With a few exceptions, we can see that speeches can be grouped differently with cluster 0 representing our 1800-1830 time period , cluster 2 representing the 1925-1950 period, cluster 1 representing the 1980-2015 period.\n",
    "\n",
    "We found it interesting how Franklin D. Roosevlet's third term was grouped by itself in cluster 3. If we take into context the year, we realize that this is only a few months after the bombing of Pearl Harbor and U.S.'s entry into World War II. Surely, this could have made a grave difference in the way his speech was written, as all of his other terms were grouped under cluster 2 until then.\n",
    "\n",
    "We can see exactly what kind of words were used for each of these clusters:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Top 20 terms per cluster:\n",
      "----------\n",
      "Cluster 0:\n",
      "----------\n",
      " america\n",
      " today\n",
      " americans\n",
      " think\n",
      " prayer\n",
      " freedom\n",
      " man\n",
      " hand\n",
      " day\n",
      " president\n",
      " help\n",
      " federal\n",
      " back\n",
      " history\n",
      " new\n",
      " young\n",
      " let\n",
      " done\n",
      " much\n",
      " bless\n",
      "----------\n",
      "Cluster 1:\n",
      "----------\n",
      " public\n",
      " duties\n",
      " union\n",
      " powers\n",
      " foreign\n",
      " general\n",
      " interests\n",
      " without\n",
      " state\n",
      " confidence\n",
      " shall\n",
      " support\n",
      " revenue\n",
      " principles\n",
      " policy\n",
      " best\n",
      " opinion\n",
      " constitution\n",
      " national\n",
      " limits\n",
      "----------\n",
      "Cluster 2:\n",
      "----------\n",
      " economic\n",
      " upon\n",
      " peoples\n",
      " leadership\n",
      " shall\n",
      " new\n",
      " democracy\n",
      " know\n",
      " conditions\n",
      " human\n",
      " common\n",
      " first\n",
      " effective\n",
      " freedom\n",
      " action\n",
      " old\n",
      " faith\n",
      " moral\n",
      " lies\n",
      " national\n",
      "----------\n",
      "Cluster 3:\n",
      "----------\n",
      " americas\n",
      " america\n",
      " today\n",
      " americans\n",
      " work\n",
      " promise\n",
      " children\n",
      " freedom\n",
      " new\n",
      " democracy\n",
      " journey\n",
      " make\n",
      " way\n",
      " build\n",
      " god\n",
      " president\n",
      " century\n",
      " generation\n",
      " schools\n",
      " courage\n"
     ]
    }
   ],
   "source": [
    "top_n = 20 # Print out the top ten terms for each cluster\n",
    "\n",
    "print(\"Top %d terms per cluster:\" %top_n)\n",
    "\n",
    "order_centroids = km.cluster_centers_.argsort()[:, ::-1]\n",
    "terms = vectorizer.get_feature_names()\n",
    "\n",
    "for i in range( num_clusters ):\n",
    "    print(\"----------\")\n",
    "    print(\"Cluster %d:\" % i)\n",
    "    print(\"----------\")\n",
    "    for ind in order_centroids[i, :top_n]:\n",
    "        print(' %s' % terms[ind]),\n",
    "    print"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Clustering Interpretation\n",
    "\n",
    "We see that *cluster 0* (1800-1830) relates more to words that describe the government and the people. Especially interesting in this cluster is the word \"union\" which shows up no where else. Union could be refering to many things like the old government under the Articles of Confderation or the generic definition with unions of entities. This cluster also talks about \"foreign\", which is likely refering to relationships with Great Britain. It seems that this cluster aims at creating a sense of country unity. This would be important because the United States is still very new and developing its hold.\n",
    "\n",
    "For *cluster 2* (1925-1955), there are words like \"peoples\", \"national\", and \"common\" that describe a whole country and again refer to unity. But unlike cluster 1, cluster 2 has words that can be seen as describing American ideals like \"spirit\", \"democracy\", and \"purpose\". The word \"industrial\" also appears here, which is quite interesting as well. We also take note of the word \"new\" which is likely showing up here because of Franklin D. Roosevelt's (FDR) \"New Deal\" which was going on in the 1930's.\n",
    "\n",
    "For *cluster 3*, which represents FDR's fourth and last term, we see that these words are quite sentimental and powerful. For example, the words \"remember\", \"truth\", 'fight\" and \"live\" all convey strong feelings. As mentioned before, these words make sense because of the context of World War II and Pearl Harbor.\n",
    "\n",
    "For *cluster 1* (1980-2015), we see a lot describe the country and it current state like \"america\", \"today\", and \"new\". This seems quite similar to *cluster 0* in its sort of calling for unity. An interesting word that popped up here is \"work\", possibly talking about jobs and employment. It seems that a lot of the words are addressed to the audience, like \"citizens\", \"children\", \"bless\", and \"god\"."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Metric 1: Total Number of Words\n",
    "Who spoke more, and what does that imply? This metric shows us a very simple, but very important measurement. How much does each person speak in their addresses or debates? This will be used in other analyses, such as word frequency and lexical diversity."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Inaugural Addresses"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count      58.000000\n",
       "mean     2341.189655\n",
       "std      1391.688757\n",
       "min       135.000000\n",
       "25%      1438.000000\n",
       "50%      2072.500000\n",
       "75%      2899.250000\n",
       "max      8467.000000\n",
       "Name: Total Words Spoken, dtype: float64"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "inaug_dat['Word Tokens'] = inaug_word_tokens\n",
    "total_words = []\n",
    "\n",
    "for i in inaug_dat.index:\n",
    "    total_words.append(len(inaug_word_tokens[i]))\n",
    "    \n",
    "inaug_dat['Total Words Spoken'] = total_words\n",
    "inaug_dat['Total Words Spoken'].describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "President William Harrison spoke the highest number of words at 8467 words.\n",
    "President George Washington spoke the lowest number of words at 135 words."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(18,4))\n",
    "ax = sns.barplot(x = inaug_dat['Term'], \\\n",
    "                 y = inaug_dat['Total Words Spoken'],errwidth=0)\n",
    "ax.set_title('Total Words Spoken for Every Presidential Term')\n",
    "# plt.xticks(rotation=-45)\n",
    "sns.despine()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/anaconda3/lib/python3.7/site-packages/scipy/stats/stats.py:1713: FutureWarning: Using a non-tuple sequence for multidimensional indexing is deprecated; use `arr[tuple(seq)]` instead of `arr[seq]`. In the future this will be interpreted as an array index, `arr[np.array(seq)]`, which will result either in an error or a different result.\n",
      "  return np.add.reduce(sorted[indexer] * weights, axis=axis) / sumval\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = sns.distplot(inaug_dat['Total Words Spoken'],\\\n",
    "                  kde=True, vertical=False, bins=58)\n",
    "sns.despine()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Debates"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Trump spoke 23650 words\n",
      "Clinton spoke 20154 words\n"
     ]
    }
   ],
   "source": [
    "# Who spoke more?\n",
    "print(\"Trump spoke\", len(TrumpTokens), \"words\")\n",
    "print(\"Clinton spoke\", len(ClintonTokens), \"words\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Debates: Total Number of Words per Candidate')"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "tw_data = [[\"Trump\", len(TrumpTokens)], [\"Clinton\",\\\n",
    "                                         len(ClintonTokens)]] \n",
    "total_words = pd.DataFrame(tw_data, columns = [\"Candidate\",\\\n",
    "                                               \"Num_words\"])\n",
    "total_words_plot = sns.catplot(x=\"Candidate\", y=\"Num_words\",\\\n",
    "                    hue=\"Candidate\", data=total_words,\\\n",
    "                    height=6, kind=\"bar\",\\\n",
    "                    palette=(sns.color_palette(\"RdBu\", n_colors=2)))\n",
    "\n",
    "total_words_plot.fig.set_size_inches(6,4)\n",
    "\n",
    "plt.xlabel(\"Candidate\")\n",
    "plt.ylabel(\"Number of Total Words Said\")\n",
    "plt.title(\"Debates: Total Number of Words per Candidate\")\n",
    "#print(tw_data)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Comparison\n",
    "\n",
    "Since inaugural addresses and debates are fundamentally different in both purpose and function, they will not be compared by this metric. There is nothing of significant interest in looking purely at total word count across different types of speeches. This is due to the fact that contexts -- including but not limited to moderator time limits and broadcasting limitations -- differ greatly between inaugural addresses and debates."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Metric 2: Most Common Words and Frequencies"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The most common words can be used to determine the focus of a candidates speech -- or when analyzed across speeches it can be used to loosely characterize a candidate's personality or values. In inaugural addresses, we used clustering to generalize cultural values by era (such as the difference between the industrial period and the modern era). This metric will be used to aid in other metrics -- such as the narcissism index and sentiment analysis."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Inaugural Addresses"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "These are our year ranges [(1770, 1820), (1820, 1870), (1870, 1920), (1920, 1970), (1970, 2020)]\n",
      "1770 - 1820\n",
      "=============\n",
      "government : 65\n",
      "may : 51\n",
      "public : 46\n",
      "people : 45\n",
      "states : 43\n",
      "citizens : 41\n",
      "every : 41\n",
      "country : 40\n",
      "us : 37\n",
      "nations : 31\n",
      "=============\n",
      "1820 - 1870\n",
      "=============\n",
      "government : 200\n",
      "states : 155\n",
      "people : 147\n",
      "may : 132\n",
      "upon : 131\n",
      "union : 122\n",
      "constitution : 116\n",
      "shall : 110\n",
      "power : 108\n",
      "great : 104\n",
      "=============\n",
      "1870 - 1920\n",
      "=============\n",
      "government : 160\n",
      "people : 157\n",
      "upon : 145\n",
      "shall : 95\n",
      "great : 92\n",
      "states : 88\n",
      "us : 77\n",
      "must : 73\n",
      "may : 72\n",
      "public : 70\n",
      "=============\n",
      "1920 - 1970\n",
      "=============\n",
      "world : 143\n",
      "people : 118\n",
      "must : 112\n",
      "us : 102\n",
      "peace : 98\n",
      "government : 90\n",
      "nations : 82\n",
      "nation : 76\n",
      "new : 72\n",
      "upon : 69\n",
      "=============\n",
      "1970 - 2020\n",
      "=============\n",
      "us : 198\n",
      "new : 117\n",
      "america : 110\n",
      "world : 108\n",
      "must : 108\n",
      "people : 100\n",
      "nation : 97\n",
      "time : 79\n",
      "freedom : 79\n",
      "government : 79\n",
      "=============\n"
     ]
    }
   ],
   "source": [
    "yearRanges = []\n",
    "speeches = []\n",
    "dists = []\n",
    "tokenized = []\n",
    "\n",
    "for i in range( 1770, 2018, 50 ):\n",
    "    yearRanges.append( (i, i+50) )\n",
    "    speeches.append('')\n",
    "\n",
    "print( \"These are our year ranges\", yearRanges )\n",
    "\n",
    "for i, t in enumerate( inaug_database ):\n",
    "    entry = inaug_database[ t ]\n",
    "    for i, speech in enumerate( entry ):\n",
    "        valid = False\n",
    "        year = speech[ 1 ].year\n",
    "        for index,r in enumerate(yearRanges):\n",
    "            # Check if the year is within a valid range\n",
    "            if year >= r[ 0 ] and year <= r[ 1 ]:\n",
    "                valid = True\n",
    "                break\n",
    "        if ( valid ):\n",
    "                speeches[index] = speeches[index] + speech[3]\n",
    "\n",
    "# Append tokenized words to the tokenized list\n",
    "for i in range( 0, len( speeches ) ):\n",
    "    t = word_tokenize( speeches[i] )\n",
    "    actual = []\n",
    "    for word in t:\n",
    "        if word not in stop_words:\n",
    "            actual.append( word )\n",
    "    tokenized.append( actual )\n",
    "    \n",
    "for i in range( 0, len( tokenized ) ):\n",
    "    dists.append( FreqDist( tokenized[i] ) )\n",
    "    \n",
    "def compareItems(a):\n",
    "    return a[1]\n",
    "\n",
    "for index in range(0, len( dists ) ):\n",
    "    print( yearRanges[index][0], '-', yearRanges[index][1] )\n",
    "    print('=============')\n",
    "    items = list( dists[ index ].items() )\n",
    "    items = sorted( items, key=compareItems, reverse=True )\n",
    "    top10 =  items[0:10]\n",
    "    for i in range(0, 10):\n",
    "        print( top10[i][0],\":\", top10[i][1])\n",
    "    print('=============')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In the first three defined year ranges, 'government' is the most frequent word. However, in the fourth year range(1920-1970), the most frequent word is 'world'. The term 'government' becomes the sixth most frequent word. For the range 1970–2020, the most frequent word is'us', and the term 'government' falls to number ten on the list.\n",
    "\n",
    "The shift from a government-centered speech to a speech focused more on a bigger pictures suggests a shift within the presidential outlook."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Debates"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [],
   "source": [
    "# What words were spoken the most by each candidate?\n",
    "\n",
    "clintonFDist = FreqDist(ClintonNoStopTokens)\n",
    "ClintonFrequency = pd.DataFrame(list(clintonFDist.items()),\\\n",
    "                                columns = [\"Word\",\"Frequency\"])\n",
    "ClintonFrequency = ClintonFrequency.sort_values(by='Frequency',\\\n",
    "                                                ascending=False)\n",
    "ClintonFrequency = ClintonFrequency.reset_index(drop=True)\n",
    "\n",
    "trumpFDist = FreqDist(TrumpNoStopTokens)\n",
    "TrumpFrequency = pd.DataFrame(list(trumpFDist.items()), columns =\\\n",
    "                              [\"Word\",\"Frequency\"])\n",
    "TrumpFrequency = TrumpFrequency.sort_values(by='Frequency',\\\n",
    "                                            ascending=False)\n",
    "TrumpFrequency = TrumpFrequency.reset_index(drop=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "Clinton_freq = ClintonFrequency.head(10)\n",
    "C_freq_num = Clinton_freq[\"Frequency\"]\n",
    "C_freq_titles = Clinton_freq[\"Word\"]\n",
    "C_freq_graph = plt.bar(C_freq_titles, height = C_freq_num,\\\n",
    "                       color = \"darkblue\")\n",
    "\n",
    "plt.xlabel(\"Most Frequent Words\")\n",
    "plt.ylabel(\"Number of Times Said\")\n",
    "plt.title(\"Clinton Debate Word Frequency\")\n",
    "\n",
    "ax = plt.gca()\n",
    "ax.spines['top'].set_visible(False)\n",
    "ax.spines['right'].set_visible(False)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "Trump_freq = TrumpFrequency.head(10)\n",
    "T_freq_num = Trump_freq[\"Frequency\"]\n",
    "T_freq_titles = Trump_freq[\"Word\"]\n",
    "T_freq_graph = plt.bar(T_freq_titles, height = T_freq_num,\\\n",
    "                       color = \"darkred\")\n",
    "\n",
    "plt.xlabel(\"Most Frequent Words\")\n",
    "plt.ylabel(\"Number of Times Said\")\n",
    "plt.title(\"Trump Debate Word Frequency\")\n",
    "\n",
    "ax = plt.gca()\n",
    "ax.spines['top'].set_visible(False)\n",
    "ax.spines['right'].set_visible(False)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "TrumpFrequency[\"candidate\"] = \"Trump\"\n",
    "ClintonFrequency[\"candidate\"] = \"Clinton\"\n",
    "C_comb_freq = ClintonFrequency.head(10)\n",
    "T_comb_freq = TrumpFrequency.head(10)\n",
    "frames = [T_comb_freq, C_comb_freq]\n",
    "combined_freq = pd.concat(frames)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1872x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "a = sns.catplot(x=\"Word\", y=\"Frequency\", hue=\"candidate\",\\\n",
    "                data=combined_freq, kind=\"bar\", palette=\"RdBu\")\n",
    "a.fig.set_size_inches(26,8)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Trump addresses himself more and Clinton addresses trump more."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Metric 3: Average Word Length\n",
    "Average word length is a simple metric -- the number of letters in all the words of a set, divided by the number of words. However, in discussions of cultural values, word length is *perceived* to be correlated with the speaker's intelligence. Word length may also correlate with target audience -- in that longer word length corresponds to a more educated target audience, while a shorter word length corresponds to a less educated (read: more widespread) audience. Therefore, analyzing and comparing word length may show us information about the candidate -- OR it may show us information about the demographics of their voting base."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define function averageWordLen\n",
    "# Pass in list of word tokens and return the calculated \n",
    "# average word length of all the word tokens\n",
    "\n",
    "def averageWordLen(words):\n",
    "    return sum(len(word) for word in words)/len(words)\n",
    "\n",
    "# Loop through all inagural address speeches and calculate\n",
    "# the average word length using the averageWordLen function\n",
    "\n",
    "i = 0\n",
    "avgWordLenList = []\n",
    "\n",
    "while (i<58):\n",
    "    avgWordLenList.append(averageWordLen(inaug_word_tokens[i]))\n",
    "    i = i + 1\n",
    "\n",
    "# Assign the average word length to a new column in the inaug_dat dataframe\n",
    "inaug_dat['Average Word Length'] = avgWordLenList"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Inaugural Addresses"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    58.000000\n",
       "mean      4.692453\n",
       "std       0.218412\n",
       "min       4.196834\n",
       "25%       4.504316\n",
       "50%       4.741030\n",
       "75%       4.870498\n",
       "max       5.027120\n",
       "Name: Average Word Length, dtype: float64"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Look at the minimum and maximum average word lengths\n",
    "inaug_dat['Average Word Length'].describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**_Minimum and Maximum Word Length_**\n",
    "\n",
    "According to average word length of inaugural addresses, Lyndon Baines Johnson having the lowest average word length at 4.196834. Grover Cleveland has the highest average word length at 5.027120.\n",
    "\n",
    "##### Context around Lyndon B. Johnson\n",
    "The inauguration of Lyndon B. Johnson was the same day that President John F. Kennedy was assassinated. \n",
    "\n",
    "##### Context around Grover Cleveland\n",
    "The Panic of 1893 was an unfavorable start to Cleveland's term. Ten days before his inauguration, the Philadelphia and Reading Railroad went bankrupt. This was the first sign of the economic crisis that would result in a depression. We believe that Cleveland incorporated longer words into his speech in the effort to sound and appear intelligent rather than \"common\" in order to ease the public's attitude amidst the worsening economony. Longer words, even if they didn't understand the meaning, might have provided the public comfort that they were in the hands of an intelligent president.\n",
    "\n",
    "###### Resources:\n",
    "\n",
    "https://resources.saylor.org/wwwresources/archived/site/wp-content/uploads/2011/08/HIST312-10.1.2-Panic-of-1893.pdf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot the distribution of the average word lengths\n",
    "\n",
    "ax = sns.distplot(inaug_dat['Average Word Length'], kde=True,\\\n",
    "                  vertical=False, bins=58)\n",
    "sns.despine()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Distribution of the Inaugural Address Average Word Length\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA7MAAAEKCAYAAADARAL7AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvOIA7rQAAIABJREFUeJzt3Xv8ZfW8+PHXuxlEUmim0hg5x/3Xj+Krg0gXOqlMuiDkGkM6dOTaQZLjnCMO4UgqThTHpZtKpcgUjsqk23RBGKQy5Xrwk1O9f3+sNWzf2ev2nVnf76zvvJ6Px358997r896f93d/39+91nuvtdeOzESSJEmSpCFZb6YTkCRJkiSpK5tZSZIkSdLg2MxKkiRJkgbHZlaSJEmSNDg2s5IkSZKkwbGZlSRJkiQNjs2sJEmSJGlwbGYlSZIkSYNjMytJkiRJGpy5M51AV7vuumuee+65M52GJEmSJKkf0WbQ4PbM3nbbbTOdgiRJkiRphg2umZUkSZIkyWZWkiRJkjQ4vX5mNiKWA/8D3AnckZkTk5a/AHhzefN3wIGZeWWfOUmSJEmShm86TgC1Y2ZWfdD1R8BTM/NXEfEM4Fjg76YhJ0mSJEnSgM3o2Ywz879Hbl4MLJipXCRJkiRJw9H3Z2YTOC8iLouIxQ1jDwDO6TkfSZIkSdIs0Pee2e0y86aImA+cHxHXZ+ZFkwdFxI4UzeyTxz1I2QgvBli4cGGf+UqSJEmSBqDXPbOZeVP5cwVwGrDt5DER8WjgeGDPzPxFxeMcm5kTmTkxb968PlOWJEmSJA1Ab81sRGwQERuuvA7sAiybNGYhcCrwwsz8Xl+5SJIkSZJmlz4PM94UOC0iVs7zmcw8NyJeBZCZxwCHAfcHji7HrfL1PZIkSZIkTRaZOdM5dDIxMZFLly6d6TQkSZIkSf2INoP6PpuxJEmSJElrnM2sJEmSJGlwbGYlSZIkSYNjMytJkiRJGhybWUmSJEnS4NjMSpIkSZIGx2ZWkiRJkjQ4NrOSJEmSpMGxmZUkSZIkDY7NrCRJkiRpcGxmJUmSJEmDYzMrSZIkSRocm1lJkiRJ0uDYzEqSJEmSBsdmVpIkSZI0ODazkiRJkqTBsZmVJEmSJA2OzawkSZIkaXBsZiVJkiRJg2MzK0mSJEkanLl9PnhELAf+B7gTuCMzJyYtD+CDwG7AH4CXZOZ3+sxJkrRu2/3Uj3Ya/6W9D+wpE0mStDp6bWZLO2bmbRXLngE8tLz8HfDR8qckSZIkSZVm+jDjPYFPZeFiYOOI2HyGc5IkSZIkreX63jObwHkRkcDHMvPYScu3AH46cvvG8r6be85L64gLjt+9c8xOL/9SD5lIkoZk71O+2Tnm1H226yETSVKVvpvZ7TLzpoiYD5wfEddn5kUjy2NMTE6+IyIWA4sBFi5c2E+mkiRJkqTB6PUw48y8qfy5AjgN2HbSkBuBB47cXgDcNOZxjs3MicycmDdvXl/pSpIkSZIGordmNiI2iIgNV14HdgGWTRp2BvCiKDwB+E1meoixJEmSJKlWn4cZbwqcVnz7DnOBz2TmuRHxKoDMPAY4m+JreW6g+Gqel/aYjyRJkiRpluitmc3MHwKPGXP/MSPXEziorxyk1XXOx3frNP4ZB5zdUyaSJEmSRs30V/NIkiRJktRZ32czlrSO+chJf99p/EH7f7mnTCRJkjSbuWdWkiRJkjQ4g90ze+tHT+o0ft6B+/eUiSRJkiRpug22mZUkSbPfnid3+yjCF/ft9lEHSdJw2cxKa6HP/+eunWOe89Jze8hEkiRJWjv5mVlJkiRJ0uC4Z1aSpGmyxymf7DT+rH1e3FMmkiQNn82s1JPTP/GMTuOf9bJzespkON7/mW6fdTvk+X6tz+rY7fQ3d445+1nvWSNz737akZ3Gf2mvN62ReSVJ0uzhYcaSJEmSpMGxmZUkSZIkDY7NrCRJkiRpcGxmJUmSJEmD4wmg1iHLP/SsTuO3fO3pPWUiSZJmm/eedkvnmDfutVkPmUhaV9jMSvorx3+q2xmFAV7+Is8qrOm1+6kf7DT+S3sf3FMmkiRppjQ2sxGxN/AeYD4Q5SUz8z4957bWWnFMt42o+a9yI0qSJEmS1qQ2e2aPBJ6Zmdf1nYzq/ewjr+0cs8VBH+ohE0mSJEmaWW2a2Z/byEqSJtvttCM6jT97r8N6ymT67H7K8Z1jvrTPy3vIRJIkVTaz5eHFAEsj4nPA6cDtK5dn5qk95yZJkiRJ0lh1e2afOXL9D8AuI7cTsJmVJEmSJM2IymY2M18KEBHbZeY3R5dFxHZtJ4iIOcBS4GeZucekZQuBTwIbA3OAt2Tm2e3T13S59uhFncY/6tVn9JSJJGkq9jj5c53Gn7Xvc3vKRJKkNWO9FmM+3PK+KgcDVZ+5fRvw+czcBtgPOLrD40qSJEmS1lF1n5l9IvAkYF5EHDKy6D4Ue1EbRcQCYHfg3cAhY4Zk+XgAGwE3tXlcSZrsnz/X/ftx3/Zcvx9XkiRpqOo+M3t34N7lmA1H7v8tsG/Lxz8KeNOk+FGHA+dFxGuADYCntXxcSZKkWs865YJO40/fZ6eeMpEk9aHuM7MXAhdGxAmZ+eOuDxwRewArMvOyiNihYtjzgBMy89/LPcEnRsRWmXnXpMdaDCwGWLhwYddUJEmSJEmzTJvvmf2PiMhJ9/2G4qROH8vMP1bEbQcsiojdgPWB+0TESZm5/8iYA4BdATLzWxGxPrAJsGL0gTLzWOBYgImJicm5aC132THPbB40yeNedWYPmaw7Tjyh2yG3L3yJh9tKkiRpWNqcAOqHwO+A48rLb4GfAw8rb4+VmYdm5oLM3JLi5E4XTGpkAX4C7AwQEY+kaHpv7fg7SJIkSZLWMW32zG6TmduP3D4zIi7KzO0j4pquE0bEEcDSzDwDeD1wXES8juJkUC/JTPe86q9889g9mgeN2G7xWT1lIkmSJGlt0aaZnRcRCzPzJ/Dn74bdpFz2pzaTZOYSYEl5/bCR+6+lOBxZkiRJkqTW2jSzrwe+ERE/AAJ4MPDqiNgA+GSfyUmSJEmSNE5jM5uZZ0fEQ4FHUDSz14+c9OmoPpOTJEmSJGmcNntmAR4HbFmOf3REkJmf6i0rSZIkSZJqNDazEXEi8LfAFcCd5d0JDLaZvfWY4zuNn/eql6+xuW/56BGdxm924GHNgyRJkiRpHdNmz+wE8CjPMixJkiRJWlu0+Z7ZZcBmfSciSZIkSVJbbfbMbgJcGxGXArevvDMzF/WWlSRJkiRJNdo0s4f3nYQkSZIkSV20+WqeCyPiQcBDM/MrEXEvYE7/qUmSJEmSNF7jZ2Yj4hXAycDHyru2AE7vMylJkiRJkuq0Ocz4IGBb4BKAzPx+RMzvNStJkqR11LNPuaZzzBf2+T89ZCJJa7c2zeztmfmniAAgIuZSfM+sJEmSxtjnlKWdxp+yz0RPmUjS7NXmq3kujIh/Au4ZEU8HvgCc2W9akiRJkiRVa9PMvgW4FbgaeCVwdma+tdesJEmSJEmq0eZsxncBx5UXACLic5n53D4Tk6TpdOgXdu00/l+ffW5PmUiSJKmNNp+ZHeeJazQLSVqH7X5Gt0b6S4v+0kg/44uLO8Wes+exncZLkiStrabazEqSJEmSpsHPP3Rhp/GbvvapPWWydqlsZiPisVWLgLv1k44kSZIkaTZYcfQpncbPf/U+ncbX7Zn995pl13eaRZIkSZKkNaiymc3MHaczEUmSJEmS2mrz1TyrJSLmRMTlEXFWxfLnRMS1EXFNRHym73wkSZIkScM3HSeAOhi4DrjP5AUR8VDgUGC7zPxVRMyfhnwkSZIkSQPX657ZiFgA7A4cXzHkFcBHMvNXAJm5os98JEmSJEmzw1TOZgxAZn6nxeMfBbwJ2LBi+cPKub4JzAEOz8xzK8ZKkiRJkgS0O5vx+sAEcCXF1/I8GrgEeHLdA0fEHsCKzLwsInaomf+hwA7AAuDrEbFVZv560mMtBhYDLFy4sG5aSZIkSdI6oPIw48zcsTyj8Y+Bx2bmRGY+DtgGuKHFY28HLIqI5cBngZ0i4qRJY24EvpiZ/5uZPwK+S9HcTs7l2HL+iXnz5rX6xSRJkiRJs1ebz8w+IjOvXnkjM5cBWzcFZeahmbkgM7cE9gMuyMz9Jw07HdgRICI2oTjs+Ictc5ckSZIkraPanM34+og4HjgJSGB/irMTT0lEHAEszcwzgC8Du0TEtcCdwBsz8xdTfWxJkiRJ0pqz4iNjv2G10vyD9ugpk1W1aWZfAhxI8RU7ABcBH+0ySWYuAZaU1w8buT+BQ8qLJEmSJEmt1DazETEHOL48PPgD05OSJEmSJEn1aj8zm5l3AvMi4u7TlI8kSZIkSY3aHGa8HPhmRJwB/H7lnZn5/r6SkiRJkiSpTptm9qbysh6wYb/pSJIkaV103KkrOo1/xd7ze8pE0lA0NrOZ+U6AiNiwuJm/6z0rSZIkSZJqNDazEbEVcCJwv/L2bcCLMvOannOTJEmSpDXmlvdf22n8Zoc8qqdMtCa0Ocz4WOCQzPwaQETsABwHPKnHvCRJ0og9Tv50p/Fn7fuCnjKRJGnt0KaZ3WBlIwvFd8ZGxAY95iRJkiRJa5Wff+DyTuM3fd02fx3/wYu7xR/8hE7j10VtmtkfRsTbKQ41Btgf+FF/KUmSJEmSVK9NM/sy4J3AqeXti4CX9paRJEmSpFnrpiNv7jT+AW/avKdMNHSVzWxEbA1cmZm/Al47fSlJkiRpqvY7dXmn8Z/de8te8pCkvtXtmT0eeHBEfAf4JvDfwMWZ+dtpyUySJEmSpArrVS3IzAnggcC7gT9R7J39fkRcGRFHT1N+kiRJkiStovYzs5n5B2BJRHwbuATYDngRsOs05CZJktYCzzz51OZBI87cd++eMpEk6S/qPjP7fIrvkt0auB1Y2dA+OTNvmZ70JEmSJElTteLD53eOmf+ap/eQyZpXt2f2WOB64Bjgosz83vSkJEnD8qpTux+scsze5/aQiSRJ0+MHH/55p/F/+5pNe8pE67K6ZnYj4DEUe2cPj4iHAzcD3wK+lZkXTEN+kiRJkiStorKZzcw7ge+Ul/+IiE2BfYHXAUcAc6YlQ0mSJKlHZ37+tk7jn/mcTXrKZPos+1i3PasAW73Svatau9R9ZvbRFHtlV17uTrFX9sMUX9UjSZIkSdKMqDvM+ASKpvUc4O2Z+eNpyUiSJEmSpAZ1hxk/dk1MEBFzgKXAzzJzj4ox+wJfAB6fmUvXxLySJEmSpNlrvWmY42DguqqFEbEh8FqKr/2RJEmSJKlRr81sRCwAdgeOrxn2LuBI4I995iJJkiRJmj1aN7MRscEUHv8o4E3AXRWPuQ3wwMw8awqPLUmSJElaR9WdAAqAiHgSxZ7VewMLI+IxwCsz89UNcXsAKzLzsojYYczy9YAPAC9pkcNiYDHAwoULm4ZLkqS1xKKTz+wcc8a+z+whE0nSbNNmz+wHgL8HfgGQmVcC27eI2w5YFBHLgc8CO0XESSPLNwS2ApaUY54AnBERE5MfKDOPzcyJzJyYN29ei6klSZIkSbNZ455ZgMz8aUSM3nVni5hDgUMByj2zb8jM/UeW/wb48zdOR8SScoxnM5YkSVJrnz3lts4x++2zSfOgtdzST6zoNH7iZfN7ykSaGW32zP60PNQ4I+LuEfEGas5O3CQijoiIRVONlyRJkiSpzZ7ZVwEfBLYAbgTOAw7qMklmLgGWlNcPqxizQ5fHlCRJkiStuxqb2cy8DXjBNOQiSZIkSVIrbc5m/KExd/8GWJqZX1zzKUmSJEmSVK/NZ2bXB7YGvl9eHg3cDzggIo7qMTdJkiRJksZq85nZhwA7ZeYdABHxUYrPzT4duLrH3CRJkiRJGqtNM7sFsAHFocWU1x+QmXdGxO29ZSZJkiSt5c7/r1s7jX/68+b1lIm07mnTzB4JXFF+D2wA2wP/EhEbAF/pMTdJkiRJksZqczbjj0fE2cC2FM3sP2XmTeXiN/aZnCRJkiRJ47Q5ARTAH4GbgV8CD4mI7ftLSZIkSZKkem2+muflwMHAAuAK4AnAt4Cd+k1NkiRJkqTx2uyZPRh4PPDjzNwR2Abo9kl3SZIkSZLWoDYngPpjZv4xIoiIe2Tm9RHx8N4zkyRJ0rR682k/6zT+PXtt0VMm64ZvnNh9/9CTX+jZkKWV2jSzN0bExsDpwPkR8SvgpoYYSZIkSZJ60+ZsxnuVVw+PiK8BGwHn9pqVJEmSJEk1apvZiFgPuCoztwLIzAunJStJkiRJkmrUngAqM+8CroyIhdOUjyRJkiRJjdp8ZnZz4JqIuBT4/co7M3NRb1lJkiRJklSjTTP7zt6zkCRJkiSpgzYngLowIh4EPDQzvxIR9wLm9J+aJEmSJEnj1X5mFiAiXgGcDHysvGsLiq/pkSRJkiRpRjQ2s8BBwHbAbwEy8/vA/D6TkiRJkiSpTptm9vbM/NPKGxExF8i2E0TEnIi4PCLOGrPskIi4NiKuioivloczS5IkSZJUq00ze2FE/BNwz4h4OvAF4MwOcxwMXFex7HJgIjMfTXEo85EdHleSJEmStI5q08y+BbgVuBp4JXA28LY2Dx4RC4DdgePHLc/Mr2XmH8qbFwML2jyuJEmSJGnd1uarefYEPpWZx03h8Y8C3gRs2GLsAcA5U5hDkiRJkrSOadPMLgKOioiLgM8CX87MO5qCImIPYEVmXhYROzSM3R+YAJ5asXwxsBhg4cKFLVKWJEmSNNvc8r4fdY7Z7A0P7iETrQ0aDzPOzJcCD6H4rOzzgR9ExNjDhifZDlgUEcspmuCdIuKkyYMi4mnAW4FFmXl7RQ7HZuZEZk7MmzevxdSSJEmSpNmszWdmycz/pTgE+LPAZRSHHjfFHJqZCzJzS2A/4ILM3H90TERsQ/H9tYsyc0XH3CVJkiRJ66jGZjYido2IE4AbgH0pTua0+VQnjIgjImJRefO9wL2BL0TEFRFxxlQfV5IkSZK07mjzmdmXUOyRfWXVYcBNMnMJsKS8ftjI/U+byuNJkiRJktZtjc1sZu43ejsitgOen5kH9ZaVJEmSJEk12uyZJSK2pjj503OAHwGn9pmUJEmSJEl1KpvZiHgYxYmbngf8AvgcEJm54zTlJkmSJEnSWHV7Zq8Hvg48MzNvAIiI101LVpIkSZIk1ag7m/E+wC3A1yLiuIjYGYjpSUuSJEmSpGqVzWxmnpaZzwUeQXEm4tcBm0bERyNil2nKT5IkSZKkVTR+z2xm/j4zP52ZewALgCuAt/SemSRJkiRJFRqb2VGZ+cvM/Fhm7tRXQpIkSZIkNenUzEqSJEmStDawmZUkSZIkDY7NrCRJkiRpcGxmJUmSJEmDYzMrSZIkSRocm1lJkiRJ0uDYzEqSJEmSBsdmVpIkSZI0ODazkiRJkqTBsZmVJEmSJA2OzawkSZIkaXBsZiVJkiRJg9N7MxsRcyLi8og4a8yye0TE5yLihoi4JCK27DsfSZIkSdLwTcee2YOB6yqWHQD8KjMfAnwAeM805CNJkiRJGrhem9mIWADsDhxfMWRP4JPl9ZOBnSMi+sxJkiRJkjR8fe+ZPQp4E3BXxfItgJ8CZOYdwG+A+/eckyRJkiRp4HprZiNiD2BFZl5WN2zMfTnmsRZHxNKIWHrrrbeusRwlSZIkScPU557Z7YBFEbEc+CywU0ScNGnMjcADASJiLrAR8MvJD5SZx2bmRGZOzJs3r8eUJUmSJElD0Fszm5mHZuaCzNwS2A+4IDP3nzTsDODF5fV9yzGr7JmVJEmSJGnU3OmeMCKOAJZm5hnAx4ETI+IGij2y+013PpIkSZKk4ZmWZjYzlwBLyuuHjdz/R+DZ05GDJEmSJGn2mI7vmZUkSZIkaY2ymZUkSZIkDY7NrCRJkiRpcGxmJUmSJEmDYzMrSZIkSRocm1lJkiRJ0uDYzEqSJEmSBsdmVpIkSZI0ODazkiRJkqTBsZmVJEmSJA2OzawkSZIkaXBsZiVJkiRJg2MzK0mSJEkaHJtZSZIkSdLg2MxKkiRJkgbHZlaSJEmSNDg2s5IkSZKkwbGZlSRJkiQNjs2sJEmSJGlwbGYlSZIkSYPTWzMbEetHxKURcWVEXBMR7xwzZmFEfC0iLo+IqyJit77ykSRJkiTNHn3umb0d2CkzHwNsDewaEU+YNOZtwOczcxtgP+DoHvORJEmSJM0Sc/t64MxM4HflzbuVl5w8DLhPeX0j4Ka+8pEkSZIkzR69fmY2IuZExBXACuD8zLxk0pDDgf0j4kbgbOA1feYjSZIkSZodem1mM/POzNwaWABsGxFbTRryPOCEzFwA7AacGBGr5BQRiyNiaUQsvfXWW/tMWZIkSZI0ANNyNuPM/DWwBNh10qIDgM+XY74FrA9sMib+2MycyMyJefPm9ZytJEmSJGlt1+fZjOdFxMbl9XsCTwOunzTsJ8DO5ZhHUjSz7nqVJEmSJNXq7QRQwObAJyNiDkXT/PnMPCsijgCWZuYZwOuB4yLidRQng3pJeeIoSZIkSZIq9Xk246uAbcbcf9jI9WuB7frKQZIkSZI0O03LZ2YlSZIkSVqTbGYlSZIkSYMTQ/uIakTcCvy4ZsgmwG1TfPjViZ3JuYea90zOPdS8Z3LuoeY9k3MPNe+ZnHuoec/k3Oa97sw91Lxncu6h5j2Tcw8175mce6h5z+TcTbG3Zebkb8JZVWbOqgvFyaWmPXYm5x5q3j5nw5p7qHn7nA1r7qHm7XO27uTtczasuYeat8/ZsOYeat5Dfs5WXjzMWJIkSZI0ODazkiRJkqTBmY3N7LEzFDuTcw8175mce6h5z+TcQ817Juceat4zOfdQ857Juc173Zl7qHnP5NxDzXsm5x5q3jM591Dznsm5VzdvYIAngJIkSZIkaTbumZUkSZIkzXKzppmNiE9ExIqIWDaF2AdGxNci4rqIuCYiDu4Qu35EXBoRV5ax75zC/HMi4vKIOGsKscsj4uqIuCIilnaM3TgiTo6I68vf/YkdYh9ezrny8tuI+McO8a8rn69lEfFfEbF+h9iDy7hr2sw5rjYi4n4RcX5EfL/8ed+O8c8u578rIiY6xr63fM6viojTImLjDrHvKuOuiIjzIuIBXeYeWfaGiMiI2KTD3IdHxM9G/ua7dZ07Il4TEd8tn7sjO8z9uZF5l0fEFR1it46Ii1f+j0TEtl3yjojHRMS3yv+zMyPiPhWxY19H2tRaTWzbOquKb6y1mtjGWquKHVneVGdVczfWWt3cTXVWM2/bOquKb6y1mti2dTZ2nRMRD46IS8o6+1xE3L1D7D9ExA11f6uG+E+Xz/eyKP6H7tYh9uPlfVdFsT66d9vYkeUfjojfTSHvEyLiRyN/8607xEZEvDsivlf+LV/bIfbrI3PeFBGnd8x754j4Thn/jYh4SIfYncrYZRHxyYiYW/O8/dW2SZsaa4hvVWcVsY011hDfWGdVsSP319ZZxbyNNdYQ31hnNbGt6qwmvrHOamK71Nkq27HRchutIrbtenNcbKvts5r4Vtto42JHltWuN2vmbrWNVjV3tNs+Gzdvq/VmTXzrbbRKa+KUyGvDBdgeeCywbAqxmwOPLa9vCHwPeFTL2ADuXV6/G3AJ8ISO8x8CfAY4awq5Lwc2meJz9kng5eX1uwMbT/Fx5gC3AA9qOX4L4EfAPcvbnwde0jJ2K2AZcC9gLvAV4KFdawM4EnhLef0twHs6xj8SeDiwBJjoGLsLMLe8/p6quSti7zNy/bXAMV3mLu9/IPBliu9rHls7FXMfDryh5d9pXPyO5d/rHuXt+V3yHln+78BhHeY9D3hGeX03YEnHvL8NPLW8/jLgXRWxY19H2tRaTWzbOquKb6y1mtjGWquK7VBnVXM31lpNbGOd1eXdss6q5m6stZrYtnU2dp1D8Tq6X3n/McCBHWK3AbakYX1SE79buSyA/+o492idvZ/yf6VNbHl7AjgR+N0U8j4B2LehzqpiXwp8Clivps4atw+AU4AXdZz7e8Ajy/tfDZzQMvZJwE+Bh5X3HwEcUPO7/9W2SZsaa4hvVWcVsY011hDfWGdVsW3rrGLexhpriG+ss7q829RZzdyNdTYulmJHWZc6W6UeaLmNVhHbdr05LrbV9llNfKtttKr/AVqsN2vmPpwW22gVsW23z8bmPbK8cr1ZM3frbbSqy6zZM5uZFwG/nGLszZn5nfL6/wDXUTRcbWIzM1e+U3e38pJt546IBcDuwPGdkl5NUbzrvz3wcYDM/FNm/nqKD7cz8IPM/HGHmLnAPct36+4F3NQy7pHAxZn5h8y8A7gQ2KsuoKI29qRo5il/PqtLfGZel5nfbUq2Iva8MneAi4EFHWJ/O3JzA2pqreZ/4gPAm6YY20pF/IHAv2Xm7eWYFV3njogAnkOxMdM2NoGVe7k2oqbWKuIfDlxUXj8f2Kcitup1pLHWqmI71FlVfGOt1cQ21lrDa2ebOlud196q2MY6a5q3RZ1VxTfWWk1s2zqrWufsBJxc3l9VZ2NjM/PyzFw+br6W8WeXyxK4lPF1VhX7W/jzc35PxtfZ2NiImAO8l6LOOufd9Ps2xB4IHJGZd5XjxtVZ7bwRsSHF323sHrOa+DZ1Ni72TuD2zPxeeX9lnU3eNin/Po01VhVf5tSqzipiG2usIb6xzqpi29bZ6m7PVcQ31lnT3E11VhPfat05Jvb+tKyzGq230SZru96siG21fVYT33obrULjerMnrbbP6jStN2u03karMmua2TUlIrakePfwkg4xc8rd6iuA8zOzdSxwFEXh3tUhZlQC50XEZRGxuEPc3wC3Av9ZHhpyfERsMMUc9qND8Wbmz4D3AT8BbgZ+k5nntQxfBmwfEfePiHtRvIvzwI75AmyamTeX+dwMzJ/CY6wJLwPO6RIQxSFHPwVeABzWMXYR8LPMvLJL3Ih/KA+h+UTVYT81HgY8JYrD1C6MiMdPYf6nAD/PzO93iPlH4L3lc/Y+4NCOcy4DFpXXn02Lepv0OtKp1qbyGtQyvrHWJsd2qbXR2KnU2Zi8W9fapNhOdVbxfLWus0nxnWptUmzrOpu8zgF+APx6ZCPsRireFFjN9VVtfBSHfr4QOLdLbET8J8XRPY/gKLnaAAAJxElEQVQAPtwh9h+AM1b+f00x73eXdfaBiLhHh9i/BZ5bHhZ3TkQ8tOO8ULwR+9VJG8Bt4l8OnB0RN1I83//WJpaiCbxb/OXQy32prrPJ2yb3p2WNVcR3URnbVGN18W3qrCK2bZ1V5d1YYzXxreqsZm5oUWcV8a3qbEzsbbSvMxi/Hdt2vTnVbeA2sU3rzLHxLdebq8R2XG9W5d5mvTkutu16s+45a7PeHBe/uttos+cw4+LNOrZkCocZj8TfG7gM2HuK8RsDXwO2ajl+D+Do8voOTO0w4weUP+cDVwLbt4ybAO4A/q68/UEqDmtreJy7U7xwbdoh5r7ABcA8ineLTwf27xB/APAdir0YxwAf6FobFCvl0eW/mkpt0XAYS0PsW4HToDireJfYctmhwDvbzk2xB/wSYKPy9nLqDxeZ/JxtSnFI+XrAu4FPdHzOlwEfojhMbFuKQ83H/u41z9lHgdd3nPdDwD7l9ecAX+kY/wiKw2AuA94B/KIh/q9eR7rU2uTYLnXWEN+m1ipf/5pqbTS2a51VPGeta21MbJc6q3q+GuusYu7WtTYmtlOdlTEr1zlPAW4Yuf+BwNUtY7caua/xb9UQfxxw1BRj5wBHAy9tGbs98A3+ckhg7eGf4+amOOQ7gHtQ7P2pPDxuTOzvVtZIWfdfn8LvfM7KeumY96n8Zd39RuD4DrFPBL5O0dj+M3D5mPGrbJtQrK9b1di4+EnLK+usRWxtjbWIr6yzit/7AW3qrGretjVWE99YZy1+59o6q5m7sc5qYhvrbOQxVtmOpeV6c1zsyLIl1B9mXBfbZp1Zu/1NzXqz4nfusn02Lr7VerMittV6s+E5a7N9Nm7uTttoYx+3a8DafGE1mlmKpurLwCGrmcM7aP+5wn+leHdzOcW7hX8ATlqNuQ/vMPdmwPKR208BvjSFOfcEzusY82zg4yO3X0T5YjiF+f8FeHXX2gC+C2xeXt8c+O5UaospNrPAi4FvAfeayrzlsgc11Tt/3cz+X4p36JeXlzso9o5vNoW5G//Xxjzn5wI7jNz+ATCvw3M2F/g5sKDjvL+BP38NWQC/XY3n/GHApTWxq7yOtK21cbEd62xsfJtaq5u7qdYmx06hzprmrvt7jHu+W9VZzfPVts7Gzd2q1lr8zrV1NmnsOyg2NG/jLxvcTwS+3DL2DSO3l9PhHAyj8eX10yk/29d17vK+p9LiTd0y9h0U682VdXYXI83WFObeocPcbwCuB7Yc+Vv/puPzdX/gF8D6HZ/vN1J8rGflfQuBa6f4O+8CfH7M2HHbJp9uW2MV8SeNLK+ss7rYNjXWNHddnVXE/qpNnbWct7LGquLb1FnDc9ZYZxXxX2pTZy1/77F1VpHL4RT/X5220UZjR24vocWbwJNjabl9Vjd3eV/jNtpI7NvpsN5sMfeWHeZ+Ax22zyqes1brzYq5O22jjbt4mDF/Ps7748B1mfn+jrHzojzbWUTcE3gaxYtPo8w8NDMXZOaWFIfqXpCZ+3eYe4PysxCUhwjvQvHuSpu5bwF+GhEPL+/aGbi27dwjnkf34+N/AjwhIu5VPvc7U3xurJWImF/+XEjxTmXX+QHOoHjBovz5xSk8xpRExK7Am4FFmfmHjrGjhxgtomWtAWTm1Zk5PzO3LGvuRooT0dzScu7NR27uRctaG3E6xed2iIiH8Ze9+m09Dbg+M2/sOO9NFBsvlPN3OUR5tN7WA95GcTTAuHFVryONtbY6r0F18W1qrSa2sdbGxXaps5q5G2ut5jlrrLOG57uxzmriG2ut5nduW2fj1jnXUex527ccVlVnU15f1cVHxMuBvweel+Vn+1rGfjfKM6SWz8szx+VTEXtZZm42Umd/yMyxZ1utyXvzkbmfxfg6q3rO/lxnFH/z73WIheJN3bMy84/jcq6Jvw7YqKxtgKczZv1Z8zuvrLN7ULw2rFJnFdsmL6BFjdXEt9q2qYptU2NV8cAL29RZxdz3bVNnNXk31lhdPC3qrOH5bqyziudsT1rUWc3v3Vhn5fKq7dg2680pbwNXxbbdPquJb7PeHBf77Q7rzaq526w3q56zNuvNuue7zXqzKn61ttGA2bNnlqKhuRn4X4oiqDxz2pjYJ1Mcx30VcEV52a1l7KOBy8vYZTQcplTzODvQ8TBjis+9XllergHe2jF+a2BpmfvpwH07xt+L4h2/jabw+76T4p98GcUZAu/RIfbrFI33lcDOU6kNincrv1r+03wVuF/H+L3K67dTvBtV9Q71uNgbKM70t7LWqs52Ny72lPI5uwo4k+JEPVP6n6D+3fFxc58IXF3OfQblu6Yd4u9O8U7zMorDxHfqkjfFWSFfNYW/9ZMpDt28kuIwnsd1jD+YYgPiexSfGao6ZHXs60ibWquJbVtnVfGNtVYT21hrVbEd6qxq7sZaq4ltrLO6vFvWWdXcjbVWE9u2zsaucyjWB5eWf/MvMOY1tSb2tWWd3UGxYTH2sNWa+Dso3slf+fussh4cF0txONw3y7/1Moq9f/dpO++kMXVnM67K+4KRuU+iPPtvy9iNKfZeXU2xF+cxXfKm2Gu0a0OdVc29VznvleXj/E2H2PdSNCXfBf6xbv5y/A785dDRxhpriG9VZxWxjTVWFd+2zqrmbltnFXk31lhDfGOd1eXdps5q5m6ss5rYVnVGxXYs7dabVbGN682a2LbbZ1Xxbdabjdvu1K83q+Zus96sim2z3qzMm3brzaq5W2+jVV1W7taVJEmSJGkwPMxYkiRJkjQ4NrOSJEmSpMGxmZUkSZIkDY7NrCRJkiRpcGxmJUmSJEmDM3emE5AkaV0SESu/dgJgM+BO4Nby9raZ+acZSUySpIHxq3kkSZohEXE4xXdXvq9DzJzMvLO/rCRJGgYPM5YkaS0RES+OiEsj4oqIODoi1ouIuRHx64j454i4FNg2Im6MiHdHxMUR8e2IeGxEnBcRP4iIV8z07yFJ0nSwmZUkaS0QEVsBewFPysytKT4KtF+5eCPgO5m5bWZ+q7xveWY+AbgY+PjKWOBd05u5JEkzw8/MSpK0dnga8HhgaUQA3BP4abnsT8Bpk8afUf68Gpibmb8Hfh8Rd0XEvTPzd9OQsyRJM8ZmVpKktUMAn8jMt//VnRFzgf+Xq57k4vby510j11fedv0uSZr1PMxYkqS1w1eA50TEJlCc9TgiFs5wTpIkrbVsZiVJWgtk5tXAO4GvRMRVwHnApjOblSRJay+/mkeSJEmSNDjumZUkSZIkDY7NrCRJkiRpcGxmJUmSJEmDYzMrSZIkSRocm1lJkiRJ0uDYzEqSJEmSBsdmVpIkSZI0ODazkiRJkqTB+f/kdSkBRRKmEAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 1152x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(16,4))\n",
    "ax = sns.barplot(x=inaug_dat['Term'],\\\n",
    "                 y=inaug_dat['Average Word Length'], errwidth=0)\n",
    "ax.set(ylim=(3.75, 5.25))\n",
    "sns.despine()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Debates"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "5.603556910569106 5.924188129899216\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Debates: Average Word Length of Candidates')"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(averageWordLen(TrumpNoStopTokens), averageWordLen(ClintonNoStopTokens))\n",
    "\n",
    "d_wl_data = [[\"Trump\", averageWordLen(TrumpNoStopTokens)],\\\n",
    "             [\"Clinton\", averageWordLen(ClintonNoStopTokens)]] \n",
    "d_word_length = pd.DataFrame(d_wl_data, columns = [\"Candidate\",\\\n",
    "                                                   \"Avg_Word_Len\"])\n",
    "d_wordl_plot = sns.catplot(x = \"Candidate\", y = \"Avg_Word_Len\",\\\n",
    "                           hue = \"Candidate\", data = d_word_length,\\\n",
    "                height = 6, kind = \"bar\", palette = \\\n",
    "                           (sns.color_palette(\"RdBu\", n_colors = 2)))\n",
    "\n",
    "d_wordl_plot.fig.set_size_inches(6,4)\n",
    "\n",
    "plt.xlabel(\"Candidate\")\n",
    "plt.ylabel(\"Average Word Length\")\n",
    "plt.title(\"Debates: Average Word Length of Candidates\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### P-Value Comparative Evaluation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Trump P-Value = 1.5129917465955484e-05\n",
      "Clinton P-Value = 8.526432891033852e-09\n"
     ]
    }
   ],
   "source": [
    "def find_p_value(a, b, c):\n",
    "    z = (a-b)/(c.std())\n",
    "    pval = stats.norm.sf(z)\n",
    "    return pval\n",
    "\n",
    "# a is the sample mean (such as trump's mean word length)\n",
    "# b is the population mean (such as the mean of inaugural word lengths)\n",
    "# c is the population data (inaugural word lengths as a whole)\n",
    "\n",
    "inaug_wl = inaug_dat['Average Word Length']\n",
    "avg_wl_trump = averageWordLen(TrumpNoStopTokens)\n",
    "avg_wl_clint = averageWordLen(ClintonNoStopTokens)\n",
    "\n",
    "p_val_wl_trump = find_p_value(avg_wl_trump, mean(inaug_wl), inaug_wl)\n",
    "print(\"Trump P-Value =\",p_val_wl_trump)\n",
    "\n",
    "p_val_wl_clint = find_p_value(avg_wl_clint, mean(inaug_wl), inaug_wl)\n",
    "print(\"Clinton P-Value =\", p_val_wl_clint)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Metric 4: Narcissism Index\n",
    "Here, we set up defined statements and functions to outline was \"narcissism\" means in a data science context. This was heavily guided by Reference #7 (see above) on narcissism analysis. Narcissism is the measurement of one's self-referential language, which reflects how \"self-centered\" they are. In politics, referencing oneself is unavoidable -- a key component of a campaign or speech is the discussion of one's own policies or goals. However, recent discussions of Donald Trump suggest that he is unusually narcissistic. Our analysis attempts to explore this possibility and verify it in comparison to his opponent Hillary Clinton, as well as past presidents."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Set up the word lists that will be used to analyze the narcissism \n",
    "# index for speeches.\n",
    "\n",
    "NARWORDS = [\"i\", \"im\", \"mine\", \"me\", \"my\", \"myself\"]\n",
    "IWORDS = [\"you\", \"we\", \"us\", \"our\", \"ours\", \"your\", \"youre\", \"yours\"]\n",
    "CWORDS = [\"america\", \"us\", \"country\", \"home\", \"american\", \"nation\"]\n",
    "\n",
    "# Count the frequency of words in tokens\n",
    "def contFreq(Tokens, Words):\n",
    "    lis = []\n",
    "    for w in Tokens:\n",
    "        if w.lower() in Words:\n",
    "            lis.append(w)\n",
    "    return (len(lis)/len(Tokens))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Inaugural Addresses"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    58.000000\n",
       "mean      0.013203\n",
       "std       0.011267\n",
       "min       0.000754\n",
       "25%       0.005679\n",
       "50%       0.010325\n",
       "75%       0.016405\n",
       "max       0.066667\n",
       "Name: Narcissism Index, dtype: float64"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Narcissism_Index = []\n",
    "\n",
    "for i in inaug_dat.index:\n",
    "    Narcissism_Index.append(contFreq(inaug_word_tokens[i], \\\n",
    "                                     NARWORDS))\n",
    "    \n",
    "inaug_dat['Narcissism Index'] = Narcissism_Index\n",
    "inaug_dat['Narcissism Index'].describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "George Washington in his second inaugural address expresses the highest narcissism index at 0.066667. \n",
    "Franklin D. Roosevelt in his third inaugural address expresses the lowest narcissim index at 0.000762."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(16,4))\n",
    "ax = sns.barplot(x=inaug_dat['Term'], \\\n",
    "                 y=inaug_dat['Narcissism Index'], errwidth=0)\n",
    "#ax.set(ylim=(0, 0.07))\n",
    "sns.despine()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "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>Term</th>\n",
       "      <th>Total Words Spoken</th>\n",
       "      <th>Average Word Length</th>\n",
       "      <th>Narcissism Index</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Term</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.133553</td>\n",
       "      <td>-0.656596</td>\n",
       "      <td>-0.536195</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Total Words Spoken</th>\n",
       "      <td>-0.133553</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.332426</td>\n",
       "      <td>-0.260555</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Average Word Length</th>\n",
       "      <td>-0.656596</td>\n",
       "      <td>0.332426</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.258436</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Narcissism Index</th>\n",
       "      <td>-0.536195</td>\n",
       "      <td>-0.260555</td>\n",
       "      <td>0.258436</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                         Term  Total Words Spoken  Average Word Length  \\\n",
       "Term                 1.000000           -0.133553            -0.656596   \n",
       "Total Words Spoken  -0.133553            1.000000             0.332426   \n",
       "Average Word Length -0.656596            0.332426             1.000000   \n",
       "Narcissism Index    -0.536195           -0.260555             0.258436   \n",
       "\n",
       "                     Narcissism Index  \n",
       "Term                        -0.536195  \n",
       "Total Words Spoken          -0.260555  \n",
       "Average Word Length          0.258436  \n",
       "Narcissism Index             1.000000  "
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "inaug_dat.corr()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = sns.lineplot(x=inaug_dat['Term'], y=inaug_dat['Average Word Length'])\n",
    "sns.despine()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Debate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Trump Narcissism  0.03543340380549683\n",
      "Clinton Narcissism  0.03304554927061625\n",
      "Trump IWORDS  0.05281183932346723\n",
      "Clinton IWORDS  0.044556911779299394\n",
      "Trump CWORDS  0.007822410147991543\n",
      "Clinton CWORDS  0.010022824253249974\n"
     ]
    }
   ],
   "source": [
    "lowerTrumpTokens = [x.lower() for x in TrumpTokens]\n",
    "lowerClintonTokens = [x.lower() for x in ClintonTokens]\n",
    "\n",
    "print(\"Trump Narcissism \", contFreq(lowerTrumpTokens, NARWORDS))\n",
    "T_nar = contFreq(lowerTrumpTokens, NARWORDS)\n",
    "print(\"Clinton Narcissism \", contFreq(lowerClintonTokens, NARWORDS))\n",
    "C_nar = contFreq(lowerClintonTokens, NARWORDS)\n",
    "\n",
    "print(\"Trump IWORDS \", contFreq(lowerTrumpTokens, IWORDS))\n",
    "T_iword = contFreq(lowerTrumpTokens, IWORDS)\n",
    "print(\"Clinton IWORDS \", contFreq(lowerClintonTokens, IWORDS))\n",
    "C_iword = contFreq(lowerClintonTokens, IWORDS)\n",
    "\n",
    "print(\"Trump CWORDS \", contFreq(lowerTrumpTokens, CWORDS))\n",
    "T_cword = contFreq(lowerTrumpTokens, CWORDS)\n",
    "print(\"Clinton CWORDS \", contFreq(lowerClintonTokens, CWORDS))\n",
    "C_cword = contFreq(lowerTrumpTokens, CWORDS)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Debates: Country-Oriented Words of Candidate')"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Narcissism\n",
    "d_nar_data = [[\"Trump\", T_nar], [\"Clinton\", C_nar]] \n",
    "d_narcissism = pd.DataFrame(d_nar_data,\\\n",
    "                            columns = [\"Candidate\", \"Narcissism\"])\n",
    "lex_diversity_plot = sns.catplot(x=\"Candidate\", y=\"Narcissism\",\\\n",
    "                                 hue=\"Candidate\", data = d_narcissism,\n",
    "                height=6, kind=\"bar\", palette=\\\n",
    "                                 (sns.color_palette(\"RdBu\", n_colors=2)))\n",
    "\n",
    "plt.xlabel(\"Candidate\")\n",
    "plt.ylabel(\"Narcissism Score\")\n",
    "plt.title(\"Debates: Narcissism of Candidate\")\n",
    "\n",
    "# Inclusivity\n",
    "d_iword_data = [[\"Trump\", T_iword], [\"Clinton\", C_iword]] \n",
    "d_iwords = pd.DataFrame(d_iword_data, columns =\\\n",
    "                        [\"Candidate\", \"IWords\"])\n",
    "d_iword_plot = sns.catplot(x=\"Candidate\", y=\"IWords\",\\\n",
    "                           hue=\"Candidate\", data = d_iwords,\n",
    "                height=6, kind=\"bar\",\\\n",
    "                           palette=(sns.color_palette(\"RdBu\", n_colors=2)))\n",
    "\n",
    "plt.xlabel(\"Candidate\")\n",
    "plt.ylabel(\"Inclusive Word Score\")\n",
    "plt.title(\"Debates: Inclusive Words of Candidate\")\n",
    "\n",
    "# Country-Oriented\n",
    "d_cword_data = [[\"Trump\", T_cword], [\"Clinton\", C_cword]] \n",
    "d_cwords = pd.DataFrame(d_cword_data, columns =\\\n",
    "                        [\"Candidate\", \"CWords\"])\n",
    "lex_diversity_plot = \\\n",
    "sns.catplot(x=\"Candidate\",\\\n",
    "            y=\"CWords\", hue=\"Candidate\", data = d_cwords,\\\n",
    "                height=6, kind=\"bar\", \\\n",
    "            palette=(sns.color_palette(\"RdBu\", n_colors=2)))\n",
    "\n",
    "plt.xlabel(\"Candidate\")\n",
    "plt.ylabel(\"Country-Oriented Word Score\")\n",
    "plt.title(\"Debates: Country-Oriented Words of Candidate\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Comparison"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Trump p-Value = 0.02424753094146427\n",
      "Clinton p-Value = 0.03911307759613424\n"
     ]
    }
   ],
   "source": [
    "inaug_nar_set = inaug_dat['Narcissism Index']\n",
    "\n",
    "nar_test_trump = find_p_value(T_nar, mean(inaug_nar_set), \\\n",
    "                              inaug_nar_set)\n",
    "print(\"Trump p-Value =\",nar_test_trump)\n",
    "\n",
    "nar_test_clint = find_p_value(C_nar, mean(inaug_nar_set), \\\n",
    "                              inaug_nar_set)\n",
    "print(\"Clinton p-Value =\", nar_test_clint)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Metric 5: Lexical Density"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Lexical Density is also known as vocabulary diversity -- it defines how many different unique words the speaker uses. It can be used to measure the intelligence of the candidate -- as well as the creativity of their campaign (or speechwriter). Higher lexical density corresponds to lower vocabulary diversity -- meaning that the speaker uses fewer unique words. In contrast, lower density means higher diversity, which means more unique words used in the speech."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Inaugural Addresses"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[596, 90, 794, 681, 778, 521, 519, 982, 1196, 966, 502, 475, 1330, 1817, 1263, 483, 1117, 894, 1014, 338, 469, 520, 807, 972, 645, 1302, 794, 1192, 812, 383, 1374, 629, 524, 1120, 1157, 1025, 709, 691, 504, 264, 743, 852, 591, 544, 531, 719, 517, 492, 846, 863, 748, 598, 724, 585, 724, 901, 786, 544]\n",
      "[0.41620111731843573, 0.6666666666666666, 0.3420939250323137, 0.3940972222222222, 0.35902168897092757, 0.44340425531914895, 0.4282178217821782, 0.29061852619118084, 0.26768128916741274, 0.33150308853809196, 0.44582593250444047, 0.4028837998303647, 0.3405889884763124, 0.21459785047832763, 0.2612202688728025, 0.44149908592321757, 0.3338314405260012, 0.31313485113835376, 0.27887788778877887, 0.4774011299435028, 0.41651865008880995, 0.3889304412864622, 0.3251410153102337, 0.32475776812562646, 0.3807556080283353, 0.29597635826324165, 0.3915187376725838, 0.2996480643539467, 0.3642889187976671, 0.38647830474268413, 0.25234159779614324, 0.36676384839650145, 0.34270765206017006, 0.33432835820895523, 0.2853267570900123, 0.27246145667198296, 0.3771276595744681, 0.3788377192982456, 0.38009049773755654, 0.48, 0.32673702726473175, 0.3481814466693911, 0.360146252285192, 0.4032616753150482, 0.3573351278600269, 0.338991041961339, 0.28674431503050474, 0.39967506092607635, 0.34474327628361856, 0.3359283768003114, 0.3200684638425332, 0.3765743073047859, 0.3372147182114579, 0.3697850821744627, 0.35179786200194363, 0.37793624161073824, 0.3766171538092956, 0.37362637362637363]\n"
     ]
    }
   ],
   "source": [
    "UniqueWords = []\n",
    "\n",
    "for i in inaug_word_tokens:\n",
    "    UniqueWords.append(list(set(i)))\n",
    "\n",
    "NumUnique = []\n",
    "for i in UniqueWords:\n",
    "    NumUnique.append(len(i))\n",
    "\n",
    "print(NumUnique)\n",
    "\n",
    "def lexical_diversity(text):\n",
    "    return len(set(text)) / len(text)\n",
    "\n",
    "LexiDiv = []\n",
    "for i in inaug_word_tokens:\n",
    "    LexiDiv.append(lexical_diversity(i))\n",
    "\n",
    "print(LexiDiv)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "inaug_dat['NumUniqueWords'] = NumUnique\n",
    "inaug_dat['LexicalDiversity'] = LexiDiv"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(16,4))\n",
    "ax = sns.barplot(x=inaug_dat['Term'], y=inaug_dat['LexicalDiversity'],\\\n",
    "                 errwidth=0)\n",
    "ax.set(ylim=(0, 0.8))\n",
    "sns.despine()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Debates"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Clinton:  2451 , Trump:  2045\n",
      "lexical diversity for Clinton is 0.12161357546888955\n",
      "lexical diversity for Trump is 0.08646934460887949\n"
     ]
    }
   ],
   "source": [
    "# Who uses more unique words?\n",
    "\n",
    "ClintonUnique = list(set(ClintonTokens))\n",
    "TrumpUnique = list(set(TrumpTokens))\n",
    "ClintonNumUnique = len(ClintonUnique)\n",
    "TrumpNumUnique = len(TrumpUnique)\n",
    "\n",
    "print(\"Clinton: \", ClintonNumUnique, \", Trump: \", TrumpNumUnique)\n",
    "\n",
    "def lexical_diversity(text):\n",
    "    return len(set(text)) / len(text)\n",
    "\n",
    "ClintonLexiDiv = lexical_diversity(ClintonTokens)\n",
    "TrumpLexiDiv = lexical_diversity(TrumpTokens)\n",
    "print(\"lexical diversity for Clinton is\", ClintonLexiDiv)\n",
    "print(\"lexical diversity for Trump is\", TrumpLexiDiv)\n",
    "\n",
    "# Clinton uses more unique words than Trump."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[['Trump', 0.08646934460887949], ['Clinton', 0.12161357546888955]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ld_data = [[\"Trump\", TrumpLexiDiv], [\"Clinton\", ClintonLexiDiv]] \n",
    "lex_diversity = pd.DataFrame(ld_data, columns = [\"Candidate\",\\\n",
    "                                                 \"Lex_div\"])\n",
    "lex_diversity_plot = sns.catplot(x=\"Candidate\", y=\"Lex_div\",\\\n",
    "                                 hue=\"Candidate\",\\\n",
    "                                 data = lex_diversity,\n",
    "                height=6, kind=\"bar\", palette=\\\n",
    "                                 (sns.color_palette(\"RdBu\",\\\n",
    "                                                    n_colors=2)))\n",
    "\n",
    "plt.xlabel(\"Candidate\")\n",
    "plt.ylabel(\"Lexical Diversity\")\n",
    "plt.title(\"Debates: Lexical Diversity per Candidate\")\n",
    "print(ld_data)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Comparison"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Metric 6: Sentiment Analysis\n",
    "Sentiment analysis is the exploration of emotionality within speech -- and in a simplified positive/negative framework can be used to characterize the speaker's attitudes or values. A more positive sentiment score (a larger score) may correspond to more optimistic language and messages of hope and triumph, while a smaller sentiment score may correspond to messages of fear and conflict. Since Donald Trump's campaign has been highly criticized for its divisive tactics, this metric is very important to understand."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Set Up"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "analyser = SentimentIntensityAnalyzer()\n",
    "\n",
    "def sentiment_analyzer_scores(sentence):\n",
    "    score = analyser.polarity_scores(sentence)\n",
    "    return score[\"compound\"]\n",
    "\n",
    "def getAvgSentiment(whatsaid):\n",
    "    totalNum = 0\n",
    "    totalScore = 0\n",
    "    for sentence in whatsaid:\n",
    "        totalScore = totalScore +\\\n",
    "        sentiment_analyzer_scores(sentence)\n",
    "        totalNum = totalNum + 1\n",
    "    return (totalScore / totalNum)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Inaugural Addresses"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "inaug_polarity = pd.DataFrame(index=inaug_dat.index,\\\n",
    "                              columns=['Positive Count',\\\n",
    "                                       'Negative Count',\\\n",
    "                                       'Neutral Count',\\\n",
    "                                       'Total Count',\\\n",
    "                                       'Positive Percent',\\\n",
    "                                       'Negative Percent',\\\n",
    "                                       'Neutral Percent'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [],
   "source": [
    "inaug_blob = TextBlob(inaug_sent[57])\n",
    "inaug_blob.sentences\n",
    "pos_count = 0\n",
    "neg_count = 0\n",
    "neu_count = 0\n",
    "\n",
    "\n",
    "for x in inaug_sent:\n",
    "    for i in range(len(inaug_sent)):\n",
    "        pos_count = 0\n",
    "        neg_count = 0\n",
    "        neu_count = 0\n",
    "        inaug_blob = TextBlob(inaug_sent[i])\n",
    "        for sentence in inaug_blob.sentences:\n",
    "            if sentence.sentiment.polarity > 0:\n",
    "                pos_count = pos_count + 1\n",
    "            elif sentence.sentiment.polarity == 0:\n",
    "                neu_count = neu_count + 1\n",
    "            else:\n",
    "                neg_count = neg_count + 1\n",
    "        inaug_polarity.iloc[i,0] = pos_count\n",
    "        inaug_polarity.iloc[i,1] = neg_count\n",
    "        inaug_polarity.iloc[i,2] = neu_count"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [],
   "source": [
    "for i in inaug_polarity.index:\n",
    "    total = 0\n",
    "    total = inaug_polarity.iloc[i,0] + inaug_polarity.iloc[i,1] +\\\n",
    "    inaug_polarity.iloc[i,2]\n",
    "    inaug_polarity.iloc[i,3] = total\n",
    "    \n",
    "    \n",
    "for i in inaug_polarity.index:\n",
    "    pos_percent = (inaug_polarity.iloc[i,0]/\\\n",
    "                   inaug_polarity.iloc[i,3])*100\n",
    "    inaug_polarity.iloc[i,4] = pos_percent\n",
    "    neg_percent = (inaug_polarity.iloc[i,1]/\\\n",
    "                   inaug_polarity.iloc[i,3])*100\n",
    "    inaug_polarity.iloc[i,5] = neg_percent\n",
    "    neu_percent = (inaug_polarity.iloc[i,2]/\\\n",
    "                   inaug_polarity.iloc[i,3])*100\n",
    "    inaug_polarity.iloc[i,6] = neu_percent"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1728x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ind = np.arange(58)\n",
    "width = 0.8\n",
    "plt.figure(figsize=(24,5))\n",
    "\n",
    "p1 = plt.bar(ind, inaug_polarity['Positive Percent'],\\\n",
    "             width, color='lightblue')\n",
    "p2 = plt.bar(ind, inaug_polarity['Negative Percent'],\\\n",
    "             width, bottom = inaug_polarity['Positive Percent'],\\\n",
    "             color='crimson')\n",
    "p3 = plt.bar(ind, inaug_polarity['Neutral Percent'],\\\n",
    "             width, bottom = inaug_polarity['Negative Percent']+\\\n",
    "             inaug_polarity['Positive Percent'], color='lightgray')\n",
    "\n",
    "plt.ylabel('Percentage')\n",
    "plt.title('Sentiment Analysis on Inaugural Addresses')\n",
    "plt.xticks(ind, inaug_dat['Term'])\n",
    "plt.yticks(np.arange(0,110,10))\n",
    "plt.legend((p1[0], p2[0]), ('Pos', 'Neg'))\n",
    "sns.despine()\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(16,4))\n",
    "ax = sns.barplot(x=inaug_polarity.index,\\\n",
    "                 y=inaug_polarity['Positive Percent'], errwidth=0)\n",
    "sns.despine()\n",
    "ax.set_title('Percentage of Positive Sentences'\\\n",
    "             ' in every inaugural address')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(16,4))\n",
    "ax = sns.barplot(x=inaug_polarity.index, y=\\\n",
    "                 inaug_polarity['Negative Percent'], errwidth=0)\n",
    "sns.despine()\n",
    "ax.set_title(\\\n",
    "'Percentage of Negative Sentences in every inaugural address')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Debates"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Trump has sentiment score: 1.0 (higher is more positive)\n",
      "Clinton has sentiment score: 1.0 (higher is more positive)\n"
     ]
    }
   ],
   "source": [
    "TrumpAvgSentiment = getAvgSentiment(whatTrumpSaid.splitlines())\n",
    "ClintonAvgSentiment = getAvgSentiment(whatClintonSaid.splitlines())\n",
    "\n",
    "print(\"Trump has sentiment score:\" , TrumpAvgSentiment,\\\n",
    "      \"(higher is more positive)\")\n",
    "print(\"Clinton has sentiment score:\" , ClintonAvgSentiment,\\\n",
    "      \"(higher is more positive)\")\n",
    "\n",
    "## Clinton is more positive."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[['Trump', 1.0], ['Clinton', 1.0]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sa_data = [[\"Trump\", TrumpAvgSentiment],\\\n",
    "           [\"Clinton\",ClintonAvgSentiment]] \n",
    "d_sent_analysis = pd.DataFrame(sa_data, columns =\\\n",
    "                               [\"Candidate\", \"Sent_ana\"])\n",
    "d_sent_analysis_plot =\\\n",
    "sns.catplot(x=\"Candidate\", y=\"Sent_ana\", hue=\"Candidate\",\\\n",
    "            data = d_sent_analysis,\\\n",
    "            height=6, kind=\"bar\",\\\n",
    "            palette=(sns.color_palette(\"RdBu\", n_colors=2)))\n",
    "\n",
    "plt.xlabel(\"Candidate\")\n",
    "plt.ylabel(\"Sentiment Score\")\n",
    "plt.title(\"Debates: Sentiment Analysis of Candidates\")\n",
    "print(sa_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [],
   "source": [
    "index = np.arange(0,2)\n",
    "# index = ['Trump','Clinton']\n",
    "deb_polarity = pd.DataFrame(index=index, \\\n",
    "                            columns=['Positive Count',\\\n",
    "                                     'Negative Count',\\\n",
    "                                     'Neutral Count',\\\n",
    "                                     'Total Count',\\\n",
    "                                     'Positive Percent',\\\n",
    "                                     'Negative Percent',\\\n",
    "                                     'Neutral Percent'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1424"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "deb_trump_blob = TextBlob(whatTrumpSaid)\n",
    "len(deb_trump_blob.sentences)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [],
   "source": [
    "pos_count = 0\n",
    "neg_count = 0\n",
    "neu_count = 0\n",
    "\n",
    "deb_trump_blob = TextBlob(whatTrumpSaid)\n",
    "for sentence in deb_trump_blob.sentences:\n",
    "    if sentence.sentiment.polarity > 0:\n",
    "        pos_count = pos_count + 1\n",
    "    elif sentence.sentiment.polarity == 0:\n",
    "        neu_count = neu_count + 1\n",
    "    else:\n",
    "        neg_count = neg_count + 1\n",
    "deb_polarity.iloc[0,0] = pos_count\n",
    "deb_polarity.iloc[0,1] = neg_count\n",
    "deb_polarity.iloc[0,2] = neu_count"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [],
   "source": [
    "pos_count = 0\n",
    "neg_count = 0\n",
    "neu_count = 0\n",
    "\n",
    "deb_clinton_blob = TextBlob(whatClintonSaid)\n",
    "for sentence in deb_clinton_blob.sentences:\n",
    "    if sentence.sentiment.polarity > 0:\n",
    "        pos_count = pos_count + 1\n",
    "    elif sentence.sentiment.polarity == 0:\n",
    "        neu_count = neu_count + 1\n",
    "    else:\n",
    "        neg_count = neg_count + 1\n",
    "deb_polarity.iloc[1,0] = pos_count\n",
    "deb_polarity.iloc[1,1] = neg_count\n",
    "deb_polarity.iloc[1,2] = neu_count"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [],
   "source": [
    "for i in deb_polarity.index:\n",
    "    total = 0\n",
    "    total = deb_polarity.iloc[i,0] + deb_polarity.iloc[i,1] +\\\n",
    "    deb_polarity.iloc[i,2]\n",
    "    deb_polarity.iloc[i,3] = total\n",
    "    \n",
    "    \n",
    "for i in deb_polarity.index:\n",
    "    pos_percent = (deb_polarity.iloc[i,0]/\\\n",
    "                   deb_polarity.iloc[i,3])*100\n",
    "    deb_polarity.iloc[i,4] = pos_percent\n",
    "    neg_percent = (deb_polarity.iloc[i,1]/\\\n",
    "                   deb_polarity.iloc[i,3])*100\n",
    "    deb_polarity.iloc[i,5] = neg_percent\n",
    "    neu_percent = (deb_polarity.iloc[i,2]/\\\n",
    "                   deb_polarity.iloc[i,3])*100\n",
    "    deb_polarity.iloc[i,6] = neu_percent"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "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>Positive Count</th>\n",
       "      <th>Negative Count</th>\n",
       "      <th>Neutral Count</th>\n",
       "      <th>Total Count</th>\n",
       "      <th>Positive Percent</th>\n",
       "      <th>Negative Percent</th>\n",
       "      <th>Neutral Percent</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>491</td>\n",
       "      <td>262</td>\n",
       "      <td>671</td>\n",
       "      <td>1424</td>\n",
       "      <td>34.4803</td>\n",
       "      <td>18.3989</td>\n",
       "      <td>47.1208</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>344</td>\n",
       "      <td>112</td>\n",
       "      <td>316</td>\n",
       "      <td>772</td>\n",
       "      <td>44.5596</td>\n",
       "      <td>14.5078</td>\n",
       "      <td>40.9326</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Positive Count Negative Count Neutral Count Total Count Positive Percent  \\\n",
       "0            491            262           671        1424          34.4803   \n",
       "1            344            112           316         772          44.5596   \n",
       "\n",
       "  Negative Percent Neutral Percent  \n",
       "0          18.3989         47.1208  \n",
       "1          14.5078         40.9326  "
      ]
     },
     "execution_count": 91,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Index 0 is Trump\n",
    "# Index 1 is Clinton\n",
    "deb_polarity"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Ethics & Privacy"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In our project, privacy is not a huge concern. All of this data are public and freely available, so there is no risk of exposing someone's private information. In fact, knowledge of political events is for the public good, as it affects all citizens, residents, and visitors of the United States (not to mention countless people abroad).\n",
    "\n",
    "As for ethics, we are not conducting any analysis that is unusual for political discussions. However, since politics is a very critical part of daily life and often incites controversy, some may argue that there are ethical implications in any political discussion or exploration. In broader domains (outside the scope of our project), ethical concerns are crucial to political data science. If the science is improper or the data is skewed or mishandled, bias can influence the public in impactful ways. It becomes easy to lie about public attitudes or the effects of certain policies if data is handled maliciously -- especially since many average citizens don't have the time to conduct their own research or fact-checking.\n",
    "\n",
    "For this reasons, it is crucial that data is handled objectively and that ethical concerns are addressed when handling political data. We have done our utmost to maintain this level of objectivity by ensuring that our hypothesis or internal feelings about certain candidates do not bleed through into our analysis. We tried to take commonly-known metrics of speech measurement (such as TF-IDF and the narcissism index) rather than developing our own metrics that could unintentionally introduce bias."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Conclusion & Discussion"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Our initial question was: how have campaign tactics changed over time, and how do they differ across party lines? To answer this question, we explored inaugural addresses of the United States Presidents (our time-scale data) and debates during the most recent presidential election (our party line data). \n",
    "\n",
    "Within each of these datasets, we were able to find interesting insights about each of our metrics. \n",
    "\n",
    "For a refresher, here is our **original hypothesis**:\n",
    "\n",
    "> \"We think that Donald Trump's debates will have a lower average word length, higher degrees of narcissism, higher density of (fewer unique) words, extreme emotional sentiment (positive and negative). We believe this will be due to Trump's lack of political experience, as well as his reputation as a brash and decisive person.\n",
    "\n",
    ">In contrast, Hillary Clinton's debates will have a higher average word length and sentence length, lower degrees of narcissism, lower density of (more unique) words, and more neutral in emotional sentiment. We believe this will be due to Clinton's experience as a politician, as well as her reputation as an educated and calculating person.\n",
    "\n",
    ">In our time-scale analysis of inaugural addresses, we hypothesize that word frequency increases, average word length and sentence length decrease, narcissism increases, and lexical density increases over time. We believe this due to an increasing appeal to the public (including those not educated about political affairs).\n",
    "\n",
    "For each of these comparative analyses, the null hypothesis is that the debate techniques (defined by our 6 metrics) of debates are not significantly different from inaugural addresses. Our alternative hypothesis is that there is a statistically significant difference between the debate techniques and the inaugural address techniques.\"\n",
    "\n",
    "**Metric 1: Total Words**\n",
    "\n",
    "- The shorter inaugural address is George Washington's second term, and the longest (by a considerable amount) was William Harrison.\n",
    "- Trump has more total words than Clinton. \n",
    "\n",
    "**Metric 2: Word Frequency**\n",
    "\n",
    "- For most of top words that both said, Trump said it more times than Clinton\n",
    "- A notable exception is the word \"think,\" which Clinton said more.\n",
    "- We did not have distinct hypotheses about word frequency,\n",
    "\n",
    "**Metric 3: Average Word Length**\n",
    "\n",
    "- Inaugural address' average words length rose then fell over time. The average of all words is skewed left (longer words earlier, words have gotten shorter).\n",
    "- The average word length is about the same for both Trump and Clinton. Both are statistically significant compared to the inaugural data, but the non-normality of the dataset may explain this unexpected result.\n",
    "\n",
    "**Metric 4: Narcissism Index**\n",
    "\n",
    "- George Washington's first two addresses are the two most narcissistic of the inaugural addresses, by a significant margin.\n",
    "- Over time, inaugural addresses have become much less self-referential, or less narcissistic.\n",
    "- Trump's narcissism index is higher than Clinton's, which supports our hypothesis, but only by a small amount.\n",
    "- Trump has more inclusive language (\"us\", \"we\") than Clinton, and the two have about the same amount of references to country-oriented words (\"nation\", \"United States\").\n",
    "\n",
    "**Metric 5: Lexical Density**\n",
    "\n",
    "- Clinton uses more unique words (lower density, higher diversity) than Trump, which supports our hypothesis.\n",
    "\n",
    "**Metric 6: Sentiment Analysis**\n",
    "\n",
    "- In the inaugural addresses, there is much more fluctuation in negative sentences --every couple of presidents sees a spike in negative language\n",
    "- Positive language fluctuates a little, but to a much lesser degree. The first 15-20 speeches are consistently positive, but the latter half of speeches see much less positive language overall.\n",
    "- In total, Clinton has a much higher sentiment score than Trump, suggesting that her speech is much more positive or optimistic than his. This supports our hypothesis that Trump's speeches rely on more negative language, but it contradicts our hypothesis in that we believed he would be more extreme negatively and positively. Clinton surprised us with her significantly higher sentiment score.\n",
    "\n",
    "From our analysis of the inaugural addresses, we saw how different historical contexts produced different addresses. This was made evident in the clustering portion of our analysis, especially with FDR's outlier clustering that was influenced by the onset of World War II. Sometimes the addresses would convey a sense of \"duty\" and \"power\" to the people; other times it called for national unity and perserverence; and some times it would be about patriotism and national pride. Though we cannot conclude anything with certainty, we can see that the addresses would change according to different time periods so as to best reflect American sentiment and ongoing events.\n",
    "\n",
    "We attempted to cross-reference our data and compare the debates with the inaugural addresses, however, our analysis was not feasible given the data we were working with. On a conceptual level, debates and inaugural addresses are extremely different in their purpose and function. We hoped we could still find important insights from it. But each of the metrics generated inaugural data that was distinctly non-normal. Therefore, when we conducted informal Z-tests with the relevant metrics and received p-values, the non-normality rendered our p-values ineffective at accurately describing statistical significance.\n",
    "\n",
    "Overall, we feel that our data gave us important results, such as showing which presidents were more narcissistic and highlighting sentiment distribution for each of the candidates. Even though our final comparative analysis was not as successful, we are excited by the information we collected from our more successful measurements. If given the opportunity, we would go further by analyzing campaign speeches or other sorts of public speaking that could lend more insight into speech tactics. "
   ]
  }
 ],
 "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
}