Machine Learning Application for Automotive Emission Prediction
Sonali Kangralkar, Rajashri Khanai
Abstract
Over the past years, the demand for transport infrastructure and automotive has increased drastically. India is one of the world's largest market for automotive and contributes significantly towards the growing economy of the country. However, it has led to an exponential increase in emissions from the transport sector. One of the major greenhouse gas emitted from an automotive being carbon dioxide (CO2). Hence there is a need to keenly monitor the emissions to avoid violating the government-specified norms. This paper emphasizes on predicting the CO2 emissions from an automotive on a real-time basis. The core focus of this work is to build various Machine Learning models to predict CO2 emissions and perform a comparative study. Based on the R2 accuracy metric the most appropriate model for this particular application has been suggested. Further to tune, the hyperparameters of the model Grid Search method are used which results in a reduction in the training time of the model.