Forecasting Health Impacts of Air Pollution with Deep Learning Models

The aim of this study is to forecast the health effects of air pollution using many machine learning methods. The dataset used for the study includes 1000 pollution measures and health outcomes—particularly respiratory and cardiovascular disorders—from Mumbai, India's dynamic metropolitan environment. The effectiveness of the convolutional neural network (CNN), artificial neural network (ANN), recurrent neural network (RNN), support vector machine (SVM), and long short-term memory (LSTM) models is investigated in detail. Using 80% of the dataset for training and 20% for testing, these models are rigorously trained and evaluated. The study's findings point to a few fascinating trends in these models’ capacity for prediction. With an astounding accuracy of 97.44%, the LSTM model leads and demonstrates how well it can capture temporal correlations. The CNN model has a 96.5% accuracy rate, making it very good at identifying spatial patterns. The impressive 92.3% performance of the ANN model demonstrates how well it can represent intricate interactions. An accuracy of 89.3% is shown in the analysis of time series data by the RNN model. Despite having the lowest accuracy (87.6%), the SVM model offers important insights into the classification of health outcomes.