Air Pollution Monitoring and Prediction using Machine Learning Algorithms

Abstract

Air pollution is an increasing environmental concern worldwide, with hazardous impacts on the environment and human health. Monitoring and predicting air pollution levels accurately are crucial for implementing and mitigating its adverse effects. The purpose of this work is to forecast the value of the Air Quality Index using data preprocessing, feature extraction and selection, and machine learning based prediction techniques. The historical air quality data was collected and analysed from the dataset that included environmental data across multiple regions of India. Linear Regression, Decision Tree, Random Forest, XGBoost, RANSAC Regression, AdaBoost and LightGBM were used to process and forecast pollution concentrations. The Random Forest and LightGBM model had the highest prediction accuracy for Air Quality Index. Key Words: Air pollution, environment and human health, Air Quality Index, Machine Learning, comparative analysis, ML models, prediction, historical data, India, Regression