PM2.5 Forecasting
A spatio-temporal deep learning model to predict fine particulate concentrations across Indian states.
This project builds a spatio-temporal deep learning model to forecast fine particulate matter (PM2.5) concentrations across India. The model combines recurrent and convolutional neural networks with attention mechanisms to capture temporal patterns and spatial correlations between states. It uses historical pollutant measurements, meteorological data and land-use information to make five‑day ahead predictions at the state level. The forecasts help policymakers and researchers anticipate air-quality episodes and evaluate emission-control strategies across diverse regions.