Tutorial review. Qualitative analysis in the near-infrared region

Abstract

This review provides an up-to-date overview of the chemometric approaches available to users of near-infrared spectroscopy specifically for the solution of qualitative analytical problems. The problems addressed include discrimination between similar materials of different functional classes; identification of pure chemicals using a previously established spectral library; selection of optimum calibration sample sets with detection of outlying and redundant samples; and preliminary investigations into the analysis of mixtures. Among procedures discussed are discriminant methods (both supervised and unsupervised) and pattern recognition techniques.