Short Term Pollution Index Prediction Using Principles of Machine Learning

Abstract

The steep rise in pollution levels in Delhi has been an extreme cause of concern. The aim of the paper is to establish a relationship between the PM2:5 pollutant concentration levels and the meteorological parameters. The study has been carried out using the principle of Machine Learning. The historical data for January 2017 of the DTU Pollution monitoring station has been used to train the machine. The same has been tested by carrying out data splits into training and test set in appropriate ratio. Using Regression and Neural Networks the prediction model has been developed accordingly.