DIFFERENT METHODS FOR PREPARATION TO DISCRIMINATE BETWEEN WILD AND CULTIVATED CIVET COFFEE USING NEAR INFRARED SPECTROSCOPY

Resumen

Civet coffee is the world’s most expensive and rarest coffee bean. Indonesia was the first country to be identified as the origin of civet coffee. At first, civet coffee is produced spontaneously by collecting civets’ feces from coffee plantations near the forest. Due to the limited stock, the farmers began cultivating the civets to obtain safe supplies of civet coffee. Based on that, civet coffee is divided into two types, wild and cultivated civet coffee. A combination of spectrometry and chemometrics can be used to evaluate authenticity with high speed and precision. In this study, 7 samples from different regions were analyzed using NIR Spectroscopy with a variety of preparations, such as unroasted, roasted, unground, and ground. The spectroscopic data were combined with the unsupervised exploratory method, hierarchical cluster analysis (HCA) and principalcomponent analysis (PCA), as well as the supervised classification method, support vector machine and random forest. The HCA results showed a trend between roasted and unroasted; meanwhile, the PCA showed a trend based on coffee beans’ regions. Combining SVM with 5-fold cross validation successfully differentiated 87.50% in all samples, 71.43% in unground samples, 33.33% in unroasted-unground samples, and 100% in ground, roasted, roasted-unground, and roasted-ground samples. However, RF in combination with 5-Fold Cross Validation successfully differentiate 75% in all samples, 57.14% in unground samples, 85.71% in ground samples, 42.86% in unroasted samples, 66.67% in unroasted-unground samples, and 100% in all roasted, roasted-unground, unroasted-ground, and roasted-ground samples. In general, roasting and groundingsamples before analysis can improve the accuracy to differentiate wild and cultivated civet coffee using NIR Spectroscopy.