AIR QUALITY INDEX PREDICTION USING MACHINE LEARNING
AIR QUALITY INDEX PREDICTION USING MACHINE LEARNING
Authors:
Vemula Lokesh1, Kalutla Sadik Hussain2, Muthyala Sneha Latha3, Banoth Murali krishna4,
Dr. Meesala Sudhir Kumar5
1234 B.Tech (AIML) Department of Computer Science, Sandip University, Nashik, Maharashtra, India.
5Department of Computer Science, Sandip University, Nashik, Maharashtra, India.
Abstract - Air pollution is one of the most critical environmental challenges, significantly affecting human health and ecosystems. The Air Quality Index (AQI) is widely used to measure pollution levels based on pollutants such as PM2.5, PM10, NO₂, SO₂, CO, and O₃. Traditional AQI estimation methods fail to capture complex, non-linear relationships between pollutants and cannot provide accurate future predictions. This study presents a Machine Learning-based AQI prediction system using regression models including Linear Regression, Decision Tree, Random Forest, and Gradient Boosting. The dataset consists of real-world air pollution data containing multiple pollutant parameters across different cities. Data pre-processing, feature engineering, and Exploratory Data Analysis (EDA) were performed to improve model performance. Among all models, the Random Forest Regressor achieved the best results with an R² score of 0.92, demonstrating superior capability in handling non-linear relationships and noisy environmental data. The system provides accurate AQI predictions and can be extended for real-time monitoring and smart city applications.
Key Words: Air Quality Index, Machine Learning, Random Forest, Pollution Prediction, Time Series Data, Environmental Monitoring.