Cleaner and faster method to detect adulteration in cassava starch using Raman spectroscopy and one-class support vector machine

摘要: Abstract Due to food adulteration concerns, analytical assays are routinely performed in labs evaluate and ensure quality control. However, classical methods used acquire reliable results lengthy costly. Therefore, we aim propose a new approach detect adulterants cassava starch clean, green, cheap, quick way. Raman spectroscopy meets all these requirements presents great potential perform such routine analyses. Data treatment is also an important step authentication problems, the use of one-class models do so. One-class support vector machine (OC-SVM) soft independent modelling by class analogy (SIMCA) were two approaches classifiers assessed this study. Cassava samples modified lab with ranging from 0.5 50%, as wheat flour, sodium bicarbonate, others. The chemometric statistically compared OC-SVM was found outperform SIMCA, reaching higher values sensitivity (87.1%), specificity (86.8%), accuracy (86.9%) prediction known data samples. This better performance resulted possibility detecting adulterations over 2% OC-SVM, only 5% SIMCA.