A Hybrid Ensemble Machine Learning Approach for Accurate Air Quality Index (AQI) Prediction in Hyderabad, Telangana
Main Article Content
Abstract
Air quality plays a crucial role in maintaining a healthy environment, yet modern industrial advances have significantly contributed to air pollution, posing various health risks. This paper presents a hybrid ensemble approach to forecast Hyderabad, Telangana's Air Quality Index (AQI) by the incorporation of machine learning techniques with advanced temporal sequence models. Specific models like Gated Recurrent Units (GRU), Prophet, and Long Short-Term Memory (LSTM) are employed to forecast AQI based on meteorological factors like temperature, wind speed, precipitation, and other relevant parameters. By applying a hybrid multivariate time series model, this study demonstrates a significant reduction in Root Mean Square Error (RMSE) in comparison to the performances of individual models, demonstrating the effectiveness of ensemble techniques in accurately predicting AQI. The results indicate that this hybrid approach is highly effective for future AQI predictions, providing a reliable tool for environmental monitoring and public health planning.