ml pca - ghdrako/doc_snipets GitHub Wiki

The main concept of PCA is the discovery of unseen relationships and correlations among attributes in the original dataset. Highly correlated attributes are so similar as to be redundant. Therefore, PCA removes such redundant attributes.