A Decision Support System for Prediction of Air Quality Using Recurrent Neural Network

When dangerous compounds are present in the air, the harmful pollutants present have an adverse effect on human health, and this is referred to as air pollution, wildlife, and the environment. Air pollution is caused by harmful substances called pollutants, which can be particulate matter, ozone, nitrogen dioxide, sulphur dioxide, and carbon monoxide. Clean, high-quality air is crucial for maintaining one’s health and well-being. This research is based on predicting the quality of air based on the number of pollutants present in the air. A particular kind of neural network called a recurrent neural network is made to process sequential input, like time series or plain language text for developing the model. Various RNN architectures can be used for air quality prediction, but LSTM is particularly effective because of its ability to capture long-term dependencies in the data and avoid vanishing gradient problems. The LSTM (univariate and multivariate) models will be used in this study. Because of its 96% accuracy, this model may be used to assess air quality. The difference between the values predicted by the values and the actual values through the forecasting can be clearly observed.