The ability of electronic nose to detect and identify volatile organic compounds makes them a promising technology for a variety of purposes. The study presents an inexpensive electronic nose system that can detect and classify Carbon-Monoxide, and Ethylene gases based on their levels of concentration. The system uses advanced algorithms to analyze and interpret the data generated by the E-nose sensors. By applying pattern recognition techniques, the electronic nose provides a more sensitive and accurate detection system. The study provides an overview of electronic nose technology and reviews various models, including gradient-boosting trees, light gradient-boosting machines, and support vector machines. The research also describes the feature extraction process, which involves principal component analysis, elastic-net, and fast Fourier transforms. The extracted features are then used to train linear regression, K-nearest neighbors, light gradient boost, and gradient boost regressor models. Overall, this research demonstrates the feasibility of developing a low-cost electronic nose system that has several applications in the real world.
