Could the quality characteristics of virgin olive oil influence its geographical classification using untargeted infrared spectroscopy and chemometrics?

Ensuring the geographical authenticity of virgin olive oil (VOO) is essential for quality control, fraud prevention, and regional product protection. This study evaluates Near-Infrared (NIR), and Infrared (IR) spectroscopies combined with chemometrics to classify 97 VOO samples from Italy (Apulia, Tuscany) and foreign countries (Morocco, Jordan, Greece, Tunisia, Spain). Partial Least Squares Discriminant Analysis (PLS-DA) effectively classified VOOs according to geographical origin, with sensitivity and specificity values higher than 0.90 in prediction. However, we also examined whether chemical quality traits, such as peroxide values and fatty acid composition, could introduce biases in the classification models. The findings suggest that inconsistent quality grade of samples could affect classification. To mitigate this, a preliminary quality assessment is recommended before applying untargeted spectroscopic methods for authentication. This study highlights the importance of integrating quality control with untargeted approaches, giving suggestions for developing more reliable authentication techniques for VOO and other foods.