作者: Carolina V. Di Anibal , Marta Odena , Itziar Ruisánchez , M. Pilar Callao
DOI: 10.1016/J.TALANTA.2009.05.023
关键词:
摘要: Abstract We propose a very simple and fast method for detecting Sudan dyes (I, II, III IV) in commercial spices, based on characterizing samples through their UV–visible spectra using multivariate classification techniques to establish rules. applied three techniques: K -Nearest Neighbour (KNN), Soft Independent Modelling of Class Analogy (SIMCA) Partial Least Squares Discriminant Analysis (PLS-DA). A total 27 spice (turmeric, curry, hot paprika mild paprika) were analysed by chromatography (HPLC-DAD) check that they free dyes. These then spiked with up concentration 5 mg L −1 . Our final data set consisted 135 distributed five classes: without dyes, I, IV. Classification results good satisfactory the mentioned above: 99.3%, 96.3% 90.4% correct PLS-DA, KNN SIMCA, respectively. It should be pointed out there are no real errors as assigned wrong class: just not any pre-defined classes.