Air Quality Prediction Using Machine Learning and Deep Learning

DOI

https://www.doi.org/10.59256/ijsreat.20260602033

Published In

Volume 6, Issue 2 (2026)

Published Online

March-April 2026

Pages

255-265

Article Type

Original Article

Check for updates

Air pollution has emerged as a critical environmental and public health issue worldwide, particularly in rapidly urbanizing and industrializing regions. The increasing concentration of harmful pollutants such as particulate matter (PM2.5 and PM10) and gaseous emissions poses serious risks to human health and the environment. Accurate prediction of the Air Quality Index (AQI) is therefore essential for effective environmental monitoring, early warning systems, and informed decision-making. However, traditional statistical models often fail to capture the complex, nonlinear, and dynamic relationships among environmental and meteorological variables, resulting in limited prediction accuracy. To address these challenges, this study proposes a hybrid framework that integrates Machine Learning (ML) and Deep Learning (DL) techniques for robust and accurate AQI prediction. The model is trained on historical air quality datasets containing key pollutant concentrations and meteorological parameters such as temperature and humidity. Multiple ML algorithms, including Logistic Regression, Support Vector Machine (SVM), Random Forest, and K-Nearest Neighbors (KNN), are implemented to establish baseline performance. Additionally, Deep Learning models such as Artificial Neural Networks (ANN) and Convolutional Neural Networks (CNN) are employed to capture complex nonlinear patterns in the data. Experimental results indicate that Logistic Regression achieved the highest accuracy of 91.44%, followed by ANN (90.75%), Random Forest (89.73%), KNN (89.04%), and CNN (89.04%), while SVM recorded an accuracy of 88.70%. These results demonstrate that machine learning models perform competitively, while deep learning models effectively capture complex data patterns. Furthermore, SHAP (Shapley Additive Explanations) is incorporated to enhance model interpretability by identifying the contribution of each feature to the prediction outcomes. The integration of explainable AI techniques ensures transparency and trust in the system. Overall, the proposed framework provides an efficient, accurate, and interpretable solution for real-world air quality prediction and environmental management applications