Prediction of Carbon Monoxide (CO) levels using machine learning and deep learning models
Constant refinement in contemporary human society has had a negative impact on air quality. Factories, vehicular engine emissions, daily necessities like gas cooking - they’ve all had no small impact in declining air quality with their emitted toxins. To combat this, there has been an increasing interest in forecasting and monitoring air quality, particularly in emerging nations like India. Among all the gaseous toxins that affect air quality, Carbon Monoxide (CO) is one of the main contributors to pollution in the air. Thus in this assignment, we will be utilising Machine Learning algorithms to predict Carbon Monoxide (CO) emission of real world places (in this case, in India) by using real world data. The data used for this study was taken from Kaggle, link to the dataset - https://www.kaggle.com/datasets/rohanrao/air-quality-data-in-india .
This study in particular will explore the best appropriate methods for foretelling Carbon Monoxide (CO) in air quality using machine learning approaches among the three models - Decision Tree Regression (DTR), Support Vector Regression (SVR), and Convolutional Neural Networks (CNN). The results produced by the following machine learning model are then evaluated using the performance metrics to find the best model. The result showed that Decision Tree Regression is the best model for this research, followed by Convolutional Neural Networks and then Support Vector Regression.