Samples of olive oils (n=67) from different qualities and samples of other vegetable edible oils (including soybean, sunflower, rapeseed, corn oil etc; n=79) were used in this study as pure oils. Previous to spectroscopy analysis, a transesterification step was applied to the pure vegetable oil samples and all the different oil blends were then prepared to create in-house blended samples. Spectral acquisition was performed with typical parameters to collect the FTIR and Raman fingerprints. For the olive/non-olive classification model, three classification strategies have been applied: (i) one inputclass (1iC) classification; (ii) two input-class (2iC) classification; and (iii) one input-class plus one 'dummy' class classification (or pseudo two input-class (p2iC) classification). The multivariate classification methods used were k-nearest neighbours (kNN), partial least squared-discriminant analysis (PLS-DA), one-class partial least squares (OCPLS), support vector machine classification (SVM-C), and soft independent modelling of class analogies (SIMCA). The multivariate quantification method used was partial least square-regression (PLS-R). FTIR fingerprints showed excellent classification ability to distinguish pure olive from non-olive oil. When PLS-DA or SVM-C techniques are applied, 100% of olive oil samples and 92% of other vegetable edible oils are correctly classified. In general FTIR fingerprints were more discriminative than Raman's in both classification and regression scenarios.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.