VARIABLE SELECTION IN DISCRIMINANT PARTIAL LEAST-SQUARES ANALYSIS

Variable selection enhances the understanding and interpretability of multivariate classification models. A new chemometric method based on the selection of the most important variables in discriminant partial least-squares (VS-DPLS) analysis is described. The suggested method is a simple extension of DPLS where a small number of elements in the we ight vector w is retained for each factor, The optimal number of DPLS factors is determined by cross-validation. The new algorithm is applie d to four different high-dimensional spectral data sets with excellent results. Spectral profiles from Fourier transform infrared spectrosco py and pyrolysis mass spectrometry are used. To investigate the unique ness of the selected variables an iterative VS-DPLS procedure is perfo rmed, At each iteration, the previously found selected variables are r emoved to see if a new VS-DPLS classification model can be constructed using a different set of variables. In this manner, it is possible to determine regions rather than individual variables that are important for a successful classification.