Please Explain Principal Component Analysis Algorithm in artificial intelligence and also write alogrithm

# Please Explain Principal Component Analysis (PCA)/SVD Algorithm in artificial intelligence

Hi!

Here, there are explanations https://github.com/masterdezign/Learning/blob/master/examples/PCA/Main.lhs

https://github.com/masterdezign/Learning/blob/master/examples/PCA2/Main.lhs

Cheers,

PCA is used to reduce dimension which is part of data preprocessing; the benefit for the PCA is reduce the dimension which will make model more stable, saving computation cost. The other benefit: since PCA is Orthogonal， it will overcome collinearity problem when you use linear algorithm and guarantee independent relationship among independent variable when you use Navie bayes algorithm

For example: you have 10 independent variable X1,X2…X10, after dimension reduction, you will have number of dimension <=10 (most case <10) depend on how much information you can squeeze from PCA.

PCA1=wX1+wX2+…wX10

PCA2=wX1+wX2+…wX10

…

When you build the model, you will use PCA instead of X1-X10, each PCA includes all X;