pca

Statistical technique to convert a set of (possibly) correlated inputs into a set of uncorrelated variables.

These uncorrelated variables are also called as principal components. PCA is performed by finding out eigen values and vectors of the covariance matrix of the input data.

PCA is also used as one of the dimensionality reduction techniques. This is done by sorting the eigen values in the descending order and only keeping the top-k eigen values and vectors. Where k is derived based on the amount of loss that is acceptable due to removal of the remaining dimensions.