Principal Component Analysis

Principal component analysis (PCA) performs a linear transformation of the coordinate system, so as to maximize the variance of the data along the first principal axis of the new coordinate system.
More information on Wikipedia.

Components Range:
You can choose the number of dimensions after projection that you keep (this might be useful to reduce the dimensionnality of the dataset for further processing)
Check the Components Range box and set the desired dimensions.

Cumulated variance and eigenvalues
The eigenvalues of each eigenvector and the cumulated variance explained by the first dimensions.

Recontruction error