Analysing and Predicting Air Quality In Delhi: Comparison of Industrial and Residential Area

Inspecting, preventing, and controlling air pollution has become one of the most essential and relevant activities in many urban and industrial cities of many nations today. Air pollution and its hazardous contents pose a threat not only to humans but also to the entire flora and fauna. With increasing levels of air pollution, we are in dire need of a model that monitors and predicts the Air quality index (PM2.5, PM10,O3, NO2, SO2, CO) over a period of time. This paper proposes a scientific contribution to this challenge. We have implemented different Machine learning algorithms like XGBoost, Random Forest, Facebook prophet, and Recurrent neural network (RNN) to forecast pollutants and particulate levels and at the same time to predict the AQI. The dataset collected is for the years 2018 and 2019 from different measurement stations in metropolitan and industrial area of Delhi city, India. The efficiency and accuracy of these supervised learning algorithms are promising and can help the meteorological department in predicting air quality.