NON-DESTRUCTIVE adulterants detection in Various honey types in Bangladesh using UV–VIS–NIR spectroscopy coupled with machine learning algorithms

Honey adulteration presents a growing challenge to food safety, product integrity, and consumer confidence. This study proposes a rapid, non-destructive approach for authenticating honey and detecting multiple adulterants using ultraviolet-visible-near infrared (UV-VIS-NIR) spectroscopy (200–900 nm) combined with machine learning algorithms—Random Forest (RF), Support Vector Machine (SVM), and Soft Independent Modeling of Class Analogy (SIMCA). Four commonly consumed honey types in Bangladesh (Wild, Blackseed, Mustard, and Rubber flower) were adulterated with corn syrup, glucose syrup, and caramel color at graded concentrations (1%, 10%, 20%, 30%). Principal Component Analysis (PCA) was used to extract informative spectral regions: 350–450 nm for botanical classification, 200–600 nm for adulteration detection, and 596–605 nm for differentiating adulterants. Among the models, RF achieved the highest performance, with classification accuracies of 100% for botanical origin, 99% for adulteration presence, and 100% for adulterant type. Physicochemical analyses supported the spectral findings, particularly in color parameters (L*, a*, b*). While further validation on larger and geographically diverse datasets is warranted, this method demonstrates strong potential as a cost-effective, scalable solution for honey authentication and food fraud detection. • UV-VIS-NIR spectroscopy (200–900 nm) was used for non-destructive honey analysis. • Four Bangladeshi honey types were tested against three common adulterants. • Machine learning models classified botanical origin, adulteration, and adulterant type. • Random Forest (RF) model showed the highest accuracy (99-100%). • The method is a rapid, scalable tool for ensuring honey authenticity in Bangladesh.