AN INTEGRATED MACHINE LEARNING FRAMEWORK FOR AIR QUALITY MONITORING AND ASSESSMENT

Air pollution poses significant risks to human health and the environment, particularly in rapidly urbanizing metropolitan regions. Accurate monitoring and prediction of air quality parameters are essential for effective policy formulation and public health interventions. This study proposes an integrated machine learning framework for monitoring and assessing air quality in the Mulund region of Mumbai using real-world sensor data collected from February 2025 till 23 January 2026. The dataset comprises pollutant concentrations including PM2.5, PM10, and O₃ obtained from the Maharashtra Pollution Control Board through the OpenAQ platform. After data preprocessing and temporal feature engineering, predictive models based on Linear Regression and Random Forest algorithms were developed to estimate PM2.5 concentrations. Experimental results demonstrate that the Random Forest model achieves superior performance with an R² value of 0.85 and a mean absolute error of 2.07 µg/m³, significantly outperforming Linear Regression. The proposed framework effectively captures nonlinear relationships and temporal patterns in air quality data, offering a reliable and scalable approach for real-time air quality assessment and decision-making in urban environments.