Evaluasi Kualitas Rasa Makanan Menggunakan Electronic Tongue

Athoillah, Ahmad (2026) Evaluasi Kualitas Rasa Makanan Menggunakan Electronic Tongue. Masters thesis, Institut Teknologi Sepuluh Nopember.

Text 6022241022-Master_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (5MB) | Request a copy

Abstract

Penilaian kualitas rasa makanan umumnya masih bergantung pada uji sensorik yang bersifat subjektif dan memerlukan waktu. Penelitian ini bertujuan mengembangkan sistem electronic tongue berbasis cyclic voltammetry untuk mengklasifikasikan kualitas rasa dan merek bumbu rawon instan menggunakan pendekatan machine learning. Sistem yang dikembangkan terdiri atas potensiostat, empat elektroda kerja (Au, Pd, Rh, dan Ir), elektroda referensi Ag/AgCl, elektroda pembantu Pt, serta perangkat lunak akuisisi dan klasifikasi data. Sebanyak enam sampel bumbu rawon diuji menggunakan sistem yang dikembangkan. Kualitas rasa ditentukan melalui uji sensorik oleh sepuluh panelis dan dikelompokkan ke dalam tiga kategori, yaitu Rendah, Sedang, dan Tinggi, sebagai ground truth. Akuisisi data menghasilkan 600 data voltammogram yang dipraolah menggunakan Moving Average dan StandardScaler. Proses klasifikasi dilakukan menggunakan Support Vector Machine (SVM) dan Random Forest (RF) dengan evaluasi Nested Leave-One-Measurement-Out Cross Validation. Hasil penelitian menunjukkan bahwa model SVM menghasilkan accuracy 81,50% dan F1-score makro 81,20% pada klasifikasi kualitas rasa, sedangkan RF menghasilkan accuracy 80,50% dan F1-score makro 80,26%. Pada klasifikasi merek bumbu rawon, model SVM dan RF masing-masing memperoleh accuracy 96,67% dan 95,00%.
===================================================================================================================================
Food taste quality assessment generally relies on sensory evaluation, which is subjective and time-consuming. This study aims to develop a cyclic voltammetry-based electronic tongue for classifying the taste quality and brand of instant rawon seasoning using a machine learning approach. The proposed system consists of a potentiostat, four working electrodes (Au, Pd, Rh, and Ir), an Ag/AgCl reference electrode, a Pt counter electrode, and integrated data acquisition and classification software. Six instant rawon seasoning samples were evaluated using the developed system. Taste quality was determined through sensory evaluation by ten panelists and categorized into three classes, namely Low, Medium, and High, which were used as the ground truth. A total of 600 voltammograms were acquired and preprocessed using Moving Average and StandardScaler. Classification was performed using Support Vector Machine (SVM) and Random Forest (RF) with Nested Leave-One-Measurement-Out Cross Validation. The results showed that the SVM model achieved an accuracy of 81.50% and a macro F1-score of 81.20% for taste quality classification, while the RF model achieved an accuracy of 80.50% and a macro F1-score of 80.26%. For seasoning brand classification, the SVM and RF models achieved accuracies of 96.67% and 95.00%, respectively.

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	pangan, electronic tongue, kualitas rasa, random forest, SVM, cyclic voltammetry, food, electronic tongue, taste quality, random forest, SVM, cyclic voltammetry
Subjects:	Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines. Q Science > QD Chemistry > QD553 Electrochemistry. Electrolysis Q Science > QD Chemistry > QD63.O9 Electrolytic oxidation. T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7871.674 Detectors. Sensors T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7878 Electronic instruments T Technology > TP Chemical technology > TP255 Electrochemistry, Industrial.
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Electrical Engineering > 20101-(S2) Master Thesis
Depositing User:	Ahmad Athoillah
Date Deposited:	28 Jul 2026 02:59
Last Modified:	28 Jul 2026 02:59
URI:	http://repository.its.ac.id/id/eprint/138054

Actions (login required)

View Item