Classification of Alcohol and Carbon Monoxide using Gas Sensors by Heuristic Algorithm based Neural Network

Abstract

The classification of electronic noses data and odors is an issue that needs to be taken to a higher level in industry, science and health. Because of Industry 4.0 and the Internet of Things is todays popular subject, it reinforces this proposal. In this study, the classification of alcohol and carbon monoxide gases which can be used frequently in industry and health fields has been classified. In order for the classification to be successful, neural networks were trained by the help of heuristic algorithms and more successful results than traditional methods have gained. Neural networks, especially trained with the Ant Colony algorithm, have achieved the best classification success in both training and test data. These results show that neural networks trained with Ant Colony algorithm will give successful results in classification of gases such as alcohol and carbon monoxide.