In recent times, the air quality level of Dhaka city has been termed as hazardous. The weather of Dhaka city has gone through some drastic changes because of extreme air pollution. Proper planning to overcome this dangerous pollution needs analysis. In this project, we have applied several machine learning models that include deep learning such as Long Short-Term Memory (LSTM) and proposed different techniques to forecast the air quality level of Dhaka city. Furthermore, we demonstrate the applicability of machine learning and deep learning models in the classification and prediction of the Air Quality Index (AQI) based on some pre-determined range. The novelty of this approach is that we have considered daily temperature as a parameter for air pollution prediction.
We conduct an extensive evaluation of these models and show that different machine learning models can classify the AQI of different places of Dhaka city. LSTM models can also forecast hourly and daily AQI with optimal performance.