Abstract:Air pollution is one of the most severe problems of the current time. It is growing day by day because of the vast level of industrialization and urbanization, causing massive damage to flora and fauna of the planet. Every moment, we are breathing air that is full of pollutants, going to our lungs, impregnating our blood and then the whole body, causing uncountable health problems. Both state and central governments have put in many efforts to keep air pollution under control. The proposed paper discusses an efficient approach towards the prediction of air quality index (AQI) of Delhi, India. AQI is a measure of air quality. It is used to inform citizens about the associated health impacts of air pollution exposure. So, we modelled a deep recurrent neural network (RNN) based on Long-Short Term Memory (LSTM) to predict hourly based concentrations of pollutants. These concentrations are then used to calculate AQI. The proposed LSTM model achieved good results in estimating hourly based ambient air quality.