dimension_reduction - JasonWayne/personal-wiki GitHub Wiki

Principal Component Analysis

PCA在寻找向量时，遵循两个原则，https://www.udacity.com/course/viewer#!/c-ud262/l-649069103/m-661438544

1. 最大化Variance
2. 正交

http://www.cs.princeton.edu/picasso/mats/PCA-Tutorial-Intuition_jp.pdf

canonical correlation analysis

Singular Value Decomposition