-
Notifications
You must be signed in to change notification settings - Fork 10
/
Copy pathch19.Rmd
49 lines (32 loc) · 1.91 KB
/
ch19.Rmd
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
---
title: "機器學習簡介"
author: "郭耀仁"
date: "`r Sys.Date()`"
output: slidy_presentation
---
```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE, warning = FALSE, results = 'hide', message = FALSE)
```
## 機器學習的定位
![](https://ingomierswacom.files.wordpress.com/2017/04/ai_ml_dl.jpg?w=676)
Source: [rapidminer](https://rapidminer.com/artificial-intelligence-machine-learning-deep-learning/)
## 定義機器學習
[Arthur Samuel](https://en.wikipedia.org/wiki/Arthur_Samuel)
> The field of study that gives computers the ability to learn without being explicitly programmed.
## 定義機器學習(2)
[Tom Mitchell](http://www.cs.cmu.edu/~tom/)
> A computer program is said to learn from experience E with respect to some class of tasks T and performance measure P, if its performance at tasks in T, as measured by P, improves with experience E.
## 定義機器學習(3)
[林軒田](http://www.csie.ntu.edu.tw/~htlin/)
> 我們從小是怎麼樣辨認一棵樹的,是爸爸媽媽告訴我們一百條規則來定義嗎?其實不是的,很大一部分是透過我們自己的觀察很多的樹與不是樹之後,得到並且內化了辨認一棵樹的技巧,機器學習想要做的就是一樣的事情。
## 機器學習的種類
- 監督式學習(Supervised Learning):具有目標變數的 Labeled data
- Gmail 會幫你分類郵件
- [Jin Yang's Hot Dog Claasifier](https://youtu.be/ACmydtFDTGs)
- 非監督式學習(Unsupervised Learning):沒有目標變數的 Unlabeled data
- 將客戶分成不同的 Segments
- 降維時所使用的主成份分析
## 機器學習的種類(2)
- 監督式學習(Supervised Learning)
- 迴歸問題:目標變數是連續型(數值),像是股價、氣溫、匯率、房價...等
- 分類問題:目標變數是離散型(類別),像是垃圾郵件/非垃圾郵件、熱狗/非熱狗...等