The classification of FT-IR spectra of different oil samples (corn,
canola, sunflower, soya, olive and butter) by means of pattern recognition
method was investigated. The principal aim of this paper
is to present chemometrics as a serious alternative to more complex
analytical tools for the analysis of ‘‘high correlated’’ data for
quality control diagnosis and, ultimately, food chemistry. PLS-DA
and iPLS-DA resulted in partial discrimination of few classes. However,
high discrimination ability was obtained by ECVA and iECVA
such that all of 255 oil samples used in this study were correctly
assigned by iECVA to their own class group with 100% sensitivity
and specificity.