Smart detection of juice adulteration: An approach based on ion mobility spectrometry and machine learning

BACKGROUND Fruit juice quality is regulated by European Union legislation; however, adulteration by adding cheaper alternatives, such as white grape juice, remains a prevalent issue. This practice constitutes economic fraud and poses health risks to consumers, including potential allergic reactions to the adulterant. Addressing this problem requires reliable methods for detecting and quantifying adulteration in fruit juices. RESULTS This study developed a method for automated data processing and quantification of adulteration levels in orange, pineapple, and apple juices adulterated with white grape juice. The approach combines headspace‐gas chromatography–ion mobility spectrometry (HS‐GC–IMS) with machine learning (ML) algorithms. Regression models, including support vector regression (SVR), partial least squares (PLS), and random forest (RF) regression, were trained. The Boruta algorithm was applied to select relevant variables, enhancing model performance. Among the models, SVR achieved the best performance with a root mean square error (RMSE) of 1.831 and a coefficient of determination ( R 2 ) of 0.987. An interactive web application was also developed to facilitate data processing and make the methodology accessible to users. CONCLUSIONS The results demonstrate the high accuracy and reliability of combining HS‐GC–IMS with ML algorithms for quantifying adulterants in fruit juices. This non‐targeted workflow, unlike traditional methods, provides an effective tool to prevent fraud and ensure consumer safety, while the web application enhances usability, allowing broader adoption in analytical settings.