Air Quality Prediction and Forecasting Using Machine Learning Algorithms: A Review

Abstract

Precise prediction of air pollutant concentrations (PM2.5, PM10, NO₂, SO₂, O₃, CO) is important for public health and environmental protection. Over the last decade, machine learning (ML) and deep learning (DL) methods have been widely used to improve the accuracy of air quality predictions. This review synthesises results from 32 recent studies, including statistical, ML, DL, hybrid, and ensemble models, such as spatiotemporal and graph-based models. It emphasizes dominant methodologies, models for prediction, challenges, and avenues for future research, highlighting the expanding involvement of interpretable and hybrid ML approaches.