COMPARISON OF CLASSIFICATION APPROACHES APPLIED TO NIR-SPECTRA OF CLINICAL-STUDY LOTS

NIR-spectroscopy combined with pattern recognition approaches is appli ed to classify samples of clinical study lots in the pharmaceutical in dustry. The performance of linear discriminant analysis (LDA), quadrat ic discriminant analysis (QDA) and K-nearest neighbour (KNN) method is evaluated on a tablet data set and a capsule data set. To establish a classification model a strategy is followed, which is described in th is work. Frequently, in the pharmaceutical industry, several batches o f the same clinical study lot are produced. We tested whether it is po ssible to merge several batches in one class for modelling or; instead , whether it is necessary to model each batch individually. (C) 1998 E lsevier Science B.V. All rights reserved.