Food frauds are characterized as changes, adulterations, manipulations and/or falsifications which are done in order to increase manufacturer’s profits. As technological progress contributes with food fraud, efficient methods are necessary to detect it. In Brazil, in accordance with current legislation (RDC no27, August 6/2010), dietary sweeteners are free of sanitary registration, which can contribute to adulterations mainly for the most expensive ones, as in the case of stévia-based sweeteners. Stévia is a natural sweetener, what turns it commercially more expensive. The aim of this study was to evaluate possible fraud in natural stevia sweeteners through spectroscopy and chemometric methods. The commercial sweeteners containing in its compositions sodium saccharin and sodium cyclamate together, stévia, sucralose and aspartame were used in this study. Their spectra were collected at UV region (200-320nm), with a 10mm quartz cuvette. A multivariate supervised classification model was constructed based on Partial Least Squares with Discriminant Analysis (PLS-DA). The scores plot shows an excellent classification of the four sweeteners groups studied. Two samples, with composition sodium saccharin, sodium cyclamate and stévia together were evaluated by the model here developed: Samples were classified in the sodium saccharin and sodium cyclamate group, indicating a low concentration of stévia in its composition. It can be concluded that UV spectroscopy coupled with the chemometric method of PLS-DA is an efficient and safe method to detect stévia sweeteners fraud.