Classification of Arabica Coffee Beans with Honey Processing Based on Starter Type Using Multi-Channel Spectra Sensor

Abstract

Arabica coffee is significant commodity in Indonesia's agricultural sector, contributing as an income source for farmers and a major export revenue. The specialty coffee industry has begun to utilize microorganisms (starter) in the fermentation process, including the honey process, to obtain a distinctive flavor. However, the use of various starters in this process results in uniform coffee bean color, making it difficult to determine the authenticity of the type of starter used. This research classifies arabica coffee beans processed with different type of starter using multi-channel spectra sensor to ensure product quality and authenticity. This research uses arabica coffee beans that are still in the form of green beans in honey processed with three types of starters, namely Saccharomyces cerevisiae, Lactobacillus sp, and Rhizopus oryzae. Multi-channel spectra sensor with PLS-DA methods can classify arabica coffee beans processed in honey with 3 types of starters. The PLS-DA model built using de-trending pre-treatment was able to predict the calibration data of arabica coffee beans with starter types very well, with accuracy, sensitivity, specificity, and precision values, reaching 97%, 95%, 96%, 95%, respectively, and the PLS-DA model built using de-trending pre-treatment is able to predict the validation sample of arabica coffee beans with starter type very well, with accuracy, sensitivity, specificity, and precision values, reaching 100%.