Characterizing Variances of Adulterated Extra Virgin Olive Oils by UV-Vis Spectroscopy Combined with Analysis of Variance-Projected Difference Resolution (ANOVA-PDR) and Multivariate Classification
Boyan Gao, Jingyao Zhang, Weiying Lu
Abstract
The analysis of variance-projected difference resolution (ANOVA-PDR) was proposed and compared with multivariate classification for its potential in detecting possible food adulteration in extra virgin olive oils (EVOOs) by UV-Vis spectra. Three factors including origin, adulteration level, and adulteration type were systematically examined by the ANOVA-derived methods. The ANOVA-PDR quantitatively presented the separation of the internal classes according to the three main factors. Specifically, the average ANOVA-derived PDRs of the EVOO origination and adulteration level, respectively, is 4.01 and 1.78, while the conventional PDRs of the three factors are all less than 1.5. Furthermore, the partial least-squares-discriminant analysis (PLS-DA) and the PLS regression (PLSR) modeling with the selected sub-datasets from different origins were used to verify the results. The resulting models suggested that the three main factors and their interactions were all important sources of spectral variations.