绪论：初识机器学习

写在前面：个人笔记整理用（实际上是用来学英语的），不具备教学功能，推荐自己去看网易云上看视频，另外推荐一个中文笔记，整理得很好http://daniellaah.github.io/2016/Machine-Learning-Andrew-Ng-My-Notes（作者记得给我广告费，滑稽）

1. 欢迎参加机器学习课程

Machine Learning

-Grew out of work in AI（源于人工智能）

-New capability for computers（计算机新的能力）

Examples:

-Datebase mining(数据挖掘)

        Large datasets from growth of automation/web.

        E.g.,Web click data, medical records, biology, engineering

-Applications can't programmby hand

        E.g., Autonomous helicopter, handwriting recognition, most of Nature Language Processing(NLP)（自然语言处理）

       Computer Vision（机器视觉）.

-Self-customizing programming（个人定制程序）

        E.g., Amazon, Netflix product recommendations

-Understanding human learning(brain, real AI)

2.什么是机器学习

Machine Learning definition:

* Arthur Samuel(1959). Machine Learning: Field of study that gives computers ability to learn without being explicitly(明确的)                      programmed.

*Tom Mitchell(1998) well-posed Learning Problem: A computer program is said to learn from experience E with respect to some task T    and some performance measure P, if its performance on T, as measured by P, immproves with experience E. 
 

Machine Lrarning algorithms:

-Supervised learning（监督学习）

-Unsupervised learning（无监督学习）

Others : Reinforcement learning（强化学习）, recommender systems（推荐系统）

3.监督学习

Supervised Learning: "right answers" given

 Example1: Hosing price prediction.

Regression【回归】: Predict continuous valued output(price)【预测连续值的输出】

 

Example2: Breast cancer(malignant, benign)【乳腺癌（恶性，良性）】

Classification【分类】: Discrete valued output(0 or 1)【输出离散值】

4.无监督学习

Clustering algorithm【聚类算法】

Examples: Google news,  Gene, Organize computing clusters, Soccial network analysis, Market segmentation, Astronomical analysis, Cocktail party problem.