MULTIWAY CALIBRATION - MULTILINEAR PLS

A new multiway regression method called N-way partial least squares (N -PLS) is presented. The emphasis is on the three-way PLS version (tri- PLS), but it is shown how to extend the algorithm to higher orders. Th e developed algorithm is superior to unfolding methods, primarily owin g to a stabilization of the decomposition. This stabilization potentia lly gives increased interpretability and better predictions. The algor ithm is fast compared with e.g. PARAFAC, because it consists of solvin g eigenvalue problems. An example of the developed algorithm taken fro m the sugar industry is shown and compared with unfold-PLS. Fluorescen ce excitation-emission matrices (EEMs) are measured on white sugar sol utions and used to predict the ash content of the sugar. The predictio ns are comparable by the two methods, but there is a clear difference in the interpretability of the two solutions. Also shown is a simulate d example of EEMs with very noisy measurements and a low relative sign al from the analyte of interest. The predictions from unfold-PLS are a lmost twice as bad as from tri-PLS despite the large number of samples (125) used in the calibration. The algorithms are available from Worl d Wide Web: hhtp:\\newton. foodsci.kvl.dk\ foodtech.