Assessing the Accuracy of Regression and Classification Models in Air Quality Prediction

This research paper addresses the pressing issue of air quality monitoring and prediction using machine learning algorithms. With a dataset comprising pollutant levels from various states, we employed a comprehensive approach to data analysis, visualization, and preprocessing. The study aimed to predict the Air Quality Index related to key pollutants such as SO2, NO2, RSPM, and SPM. We utilized linear regression, decision tree regression, and random forest regression for prediction tasks and logistic regression, decision tree classification, random forest classification, and k-nearest neighbors for classification tasks. The proposed method was evaluated using different performance matrixes for regression and classification tasks, respectively. The results demonstrated the usefulness of the proposed models in predicting and classifying air quality levels. The proposed model provides a robust framework for air quality management and can be a valuable tool for environmental monitoring and policymaking.