Principal Component Analysis – PCA in R
PCA is a mechanism of feature reduction, PCA takes into account for the maximum variance in the predictor variables and generates orthogonal projections. The principal components PC1, PC2 and so on and so forth are either less or equal to the number of attributes. Essentially they help us in determining the regressand with less number of principal components than the number of attributes. PCA is an excellent technique to reduce regressors without compromising the variance in the data. Auto-encoders for deep learning (DNN) perform the similar task. This volg introduces you to PCA with hands on R example.