Enhanced Differentiation of Wild and Feeding Civet Coffee Using Near-Infrared Spectroscopy with Various Sample Pretreatments and Chemometric Approaches

Civet coffee is the world’s most expensive and rarest bean. Indonesia was first country to be identified as origin of civet coffee. First, it produced spontaneously by collecting feces from plantations near forest. Due limited stock, farmers began cultivating civets obtain safe supplies Based on this, can divided into two types: wild fed. A combination spectroscopy chemometrics used evaluate authenticity with high speed precision. In this study, seven samples different regions were analyzed using NIR Spectroscopy various preparations: unroasted, roasted, unground, ground. The spectroscopic data combined unsupervised exploratory methods (hierarchical cluster analysis (HCA) principal component (PCA)) supervised classification (support vector machine (SVM) random forest (RF)). HCA results showed a trend between roasted unroasted beans; meanwhile, PCA based bean regions. Combining SVM leave-one-out-cross-validation (LOOCV) successfully differentiated 57.14% in all sample groups (unground, ground, unroasted–unground, roasted–unground), 78.57% 92.86% roasted–ground, 100% unroasted–ground. However, Boruta filter, accuracy increased 89.29% for samples, 85.71% unground unroasted–ground, roasted–ground. Ultimately, RF grouped samples. general, roasting grinding before improved differentiating feeding Spectroscopy.