LAUSR

Abstract Goat dairy products are the target of fraudulent practices due to their high commercial value. This study evaluated the capacity of vibrational spectroscopy Near Infrared (FT-NIR) coupled with chemometric tools to detect goat’s yogurt and cheese adulterated with cow milk. Principal Component Analysis (PCA), Q-control chart and Partial Least Squares-Discriminant Analysis (PLS-DA) were able to distinguish goat’s cheese and yogurt adulterated with 10, 15 and 20% of cow milk. The Q-control chart was able to discern between both (authentic and adulterated) evaluated products. PLS-DA showed 100% sensitivity and specificity in discriminating samples of yogurt and goat cheese. The algorithm Interval Partial Least Square (iPLS) reduced the need for 6001 variables to 140 in the yogurt model and 70 variables in the cheese model, without affecting performance quality. FT-NIR demonstrated reliability in evaluating the authenticity of goat products and future studies employing simplified equipment (portable) and on-line and/or in process applicability should be prioritized.