Machine Learning lesson 12 http://v.163.com/movie/2008/1/O/T/M6SGF6VB4_M6SGKGMOT.html

1. clustering

2. k-means clustering algorithm

K-means算法是很典型的基于距离的聚类算法，采用距离作为相似性的评价指标，即认为两个对象的距离越近，其相似度就越大。

3. density estimation

anomaly detection or standout analysis

4. EM algorithm

这种方法可以广泛地应用于处理缺损数据，截尾数据，带有噪声等所谓的不完全数据(incomplete data)

see more inhttp://baike.baidu.com/view/1541707.htm

Gaussian Discriminant Analysis

maximum likeliness estimation

Jensen's inequality:

In the context of probability theory,it is generally stated in the following form: if X is a random variable and f is a convex function,then f(EX) <= E(f(X)).