机器学习基础之PCA-协方差矩阵

基础：https://www.zybuluo.com/mdeditor#1468315-full-reader

协方差

covariance, correlation & linear regression are closely related.

covariance解释和测量linear relationship的方向，correlation解释linear relationship的大小
covariance为正，表示向同一方向move

求解协方差矩阵的步骤

c o v ( X 1 , X 2 ) = E [ ( X 1 − μ X 1 ) ( X 2 − μ X 2 )