Machine Learning-Enabled Smart Gas Sensing Platform for Carbon Monoxide Identification

This study focuses on identifying the best machine learning model for carbon monoxide (CO) detection. CO, a colorless and odorless gas, poses significant health risks, making its detection crucial in various settings. The research evaluates several machine learning algorithms, including K-Nearest Neighbors (KNN), Artificial Neural Networks (ANN), Support Vector Machines (SVM), and Multi-Layer Perceptrons (MLP), to enhance the precision of CO detection. The results indicate that ANN and SVM perform well but are prone to overfitting with complex data. MLP showed the highest error rate, suggesting the need for further optimization. This study highlights the importance of selecting appropriate machine learning models based on dataset characteristics and application context, aiming to improve environmental monitoring and proactive safety management.