<![CDATA[Coffe is one of Indonesia’s leading commodities, known for its diverse flavors and aromas. Traditionally, coffee quality assessment is conducted manually through cupping tests performed by expert panelists. However, this method is subjective and requires considerable time and cost. This study aims to implement an Artificial Neural Network (ANN) using the backpropagation algorithm to classify coffee types based on sensory parameters such as flavor, aroma, acidity level, and body. Simulated data were generated from five common Indonesian coffee varieties: Arabica Gayo, Robusta Lampung, Arabica Toraja, Liberica Jambi, and Excelsa. The results show that the ANN-based classification system with a 4-8-1 architecture achieved an accuracy rate of 93% after 500 training epochs, with a final error value of 0.07. The implementation of this method provides an efficient and objective technological alternative to assist the coffee industry in maintaining product quality and automatically identifying coffee types.]]>
Copyrights © 2025
