Urbanization and Industrialization during the last few decades have increased air pollution causing harm to human health. Air pollution in metro cities turns out to be a serious environmental problem, especially in developing countries like India. The major environmental challenge is, to predict accurate air quality from pollutants. Envisaging air quality from pollutants like PM2.5, using the latest deep learning technique (LSTM timer series) has turned out to be a significant research area. The primary goal of this research paper is to forecast near-time pollution using the LSTM time series multivariate regression technique. The air quality data from Central Pollution Control Board over Hyderabad station has been used for the present study. All the processing is done in real-time and the system is found to be functionally very stable and works under all conditions. The Root Mean Square Error (RMSE) and R2 have been used as evaluation criteria for this regression technique. Further, the time series regression has been used to find the best fit model in terms of processing time to get the lowest error rate. The statistical model based on machine learning established a relevant prediction of PM2.5 concentrations from meteorological data.

Poor air quality is becoming a critical environmental concern in different countries over the last several years. Most of the air pollutants have serious consequences on human health and wellbeing. In this context, efficient forecasting of air pollutants is currently crucial to predict future events with a view to taking corrective actions and framing effective environmental policies. Although deep learning (DL) as well as statistical forecasting models are investigated in the literature, they have rarely used in air pollutant-specific optimal model building for long-term forecasting. In this paper, our aim is to develop the pollutant-specific optimal forecasting models for the phases spanning from preprocessing to model building by investigating a set of predictive techniques. In this regard, this paper presents a methodology for long-term forecasting of some important air pollutants. More specifically, a total of eight best performing models such as stacked LSTM, LSTM auto-encoder, Bi-LSTM, convLSTM, Holt-Winters, auto-regressive (AR), SARIMA, and Prophet are investigated for developing pollutant-specific optimal forecasting models. The study is carried out based on the real-world data obtained from government-run air quality monitoring units in Kolkata over a period of 4 years. The models such as Holt-Winters, Bi-LSTM, and ConvLSTM achieve high forecasting accuracy with respect to MAE and RMSE values for majority of the pollutants.