IDENTIFIKASI AROMA CAMPURAN (BLENDING) KOPI ARABIKA DAN ROBUSTA DENGAN ELECTRONIC NOSE MENGGUNAKAN SISTEM PENGENALAN POLA Scent Identification with Electronic Nose Method on Arabica And Robusta Coffee Blending using Pattern Recognition System

Aim of this study was to determine value of Arabica and Robusta coffee blending using Neural Network pattern recognition systems based on Electronic Nose scent readings. The architecture was built using 4 pieces hidden layer with 4 neurons input in the form of Electronic Nose gas readings. While for output (target) was the value of blending coffee with three neurons which are a combination of binary digits 0 and 1. Artificial Neural Networks developed in this research can be used in recognizing the value of Arabica and Robusta coffee blending with an accuracy of 75.