Predicting air quality using intelligent techniques

Air pollution is the major concern which not only impacts the environment but also all the inhabitants of earth. Inhaling the detrimental air may lead to several diseases or even death. So, it culminates in the need of continuous monitoring of pollution levels of various pollutants such as SO2, O3, NO2, CO (PM10 and PM2.5). Proposed work not only finds patterns in the previous data but also be able to alert by predicting the concentration of these pollutants in the near future. Results show how a time series problem can be converted into a supervised problem and contrasted their performance. Comparative analysis done using Long short term memory(LSTM) with auto regressive integrated moving average (ARIMA) to predict pollutant and results proved that LSTM is a better choice.