Principal Component Analysis (PCA) allows to  reduce the number of variables of a data set, while preserving as much information as possible. Because smaller data sets are easier to explore and visualize and make analyzing data much easier and faster for machine learning algorithms without extraneous variables to process.