Air Quality Prediction Using Machine Learning

Abstract

Abstract Air pollution is one of the current major problems in the world, and due to this, the quality of air we breathe is becoming worse as the days pass. Air pollution has been increasing rapidly since the year 2010, as most of the reports say that every year since 2015 air pollution has been more than that of total air pollution recorded in the previous whole decade! So, to live a sustainable life, the quality of air we breathe must be good and free of any kind of pollutants. So, to predict and monitor the air quality the data of various air pollutants that decrease the air quality have been collected and used as features for developing a machine learning model which predicts the air quality index of a particular place given the values of the pollutants. Machine learning models like Linear regression, Logistic regression, and Artificial Neural Networks (ANNs) models have been used and compared in terms of their accuracy. Initially, simple machine learning models like linear and logistic regression were trained and achieved good accuracies, later the use of complex artificial neural networks proved to have the highest accuracy of them all on test data sets.