Prediction of Air Quality Index Using Machine Data Learning on Atmospheric

Main Article Content

M. Mallegowda, Dr. Anita Kanavalli, Yash Verma, S Jaya Krishna Vamsi, Tata Mukesh, Vedant Saxena

Abstract

Air Quality Index (AQI) is a standard measure of pollutant levels such as those of PM2.5, PM10, SO2, NO2, O3, CO, NH3 and Lead over a period. Nowadays, the atmosphere is getting polluted rapidly. So, there is an urge to know how the air is going to be around us in the near future. We are implementing an interactive and user-friendly web application, where users find out the AQI and also predict PM2.5 AQI for the next day. We have trained a XgBoost model. The application is also capable of plotting graphs of temperature, humidity and AQI. These graphs can be visualized based on monthly or weekly data. In this paper, we discuss how data was collected and cleaned, how feature engineering was applied as part of pre-processing and finally what all models we tried and came to the conclusion of XgBoost being the most suitable to our use-case.

Article Details

Section
Articles