Air Quality Index Prediction using Machine Learning Techniques
Kumaresh J
Air pollution is one of the most important health and environmental issues today. The Air Quality Index (AQI) is gaining popularity as a recognized indicator of the safety of the ambient air that is easily understood by the public and the decision-makers. This study focuses on the use of machine learning models to predict AQI values using five years of pollutant data collected from some Indian cities. The pollutants investigated are sulfur dioxide (SO₂), nitrogen dioxide (NO₂), respirable suspended particulate matter (RSPM), and suspended particulate matter (SPM). Several machine learning models were employed and studied in the analysis, including Linear Regression, Logistic Regression, Decision Trees, Random Forest, and K-Nearest Neighbors (KNN), to determine their potential for predicting AQI categories or predicting it as a continuous variable. The analysis showed that Random Forest was the most reliable and accurate approach, as it achieved the best balance between accuracy, interpretability, and generalization. This study raises awareness of the merit of developing predictive models that can and potentially lead to early warning systems and decision support systems for sustainable urban management and public health improvement.