Interpretation of infrared spectra based on statistical approaches

A computer procedure is presented for deriving classification rules automatically from a given set of i.r. spectra. The percentage of peaks belonging to a cluster within a set of spectra is used as a measure to train classification rules and define families of spectra. The algorithm obtains information from the fingerprint region of the i.r. spectra and is tested for natural products in different ways. The information is used for fast preliminary interpretation of spectra and the search time is decreased considerably. The study is based on the SPEKTREN information system of the German Cancer Research Center.