Abstract—Harmful gases can inflict damage, accident, or even death on a human being. The safety and security of human beings are essential to the community. This study aims to develop an electronic nose with prediction modeling using machine learning. With the use of the sensor substitution technique, the system can predict the gas by utilizing a minimal number of sensors. The system, which is designed to predict Ammonia, Ethanol, and Isobutylene, relies on Support Vector Machine (SVM), Gaussian Process Regression (GPR), and Feed Forward Neural Network (FFNN) to ensure the prediction models are accurate. As a result, for Ammonia, Ethanol, and Isobutylene, using SVM, GPR, and FFNN model attain value of 1.00 for R2. The RMSE and MAE has a low result value indicates that the SVM, GPR, and FFNN model is performing well and can be used to make decisions regarding the concentration levels of harmful gases. This study shows that the system can predict the presence of gases within it using machine learning.
 
Keywords—machine learning, e-nose, prediction model, support vector machine, gaussian process regression, feed forward neural network


Copyright © 2023 by the authors. This is an open access article distributed under the Creative Commons Attribution License (CC BY-NC-ND 4.0), which permits use, distribution and reproduction in any medium, provided that the article is properly cited, the use is non-commercial and no modifications or adaptations are made.