Air Quality Index Prediction using Machine Learning Techniques

Pollution is the most vital and disturbing issues faced in today’s world. Above 2000 people die due to diseases whose root cause is pollution. Pollution can be of various forms and each of these types can have different effects on different people. Increased pollution levels are capable of causing mass destruction to the earth as well as to the species residing in it. One of the prevalent environmental challenges right now is air pollution. Air pollution has been noticeable as one of the most important problems of metropolitan regions around the globe, exclusively in Delhi, Beijing and Tehran where its inhabitants and administrators have long been struggling with air pollution impairment such as the health issues of its citizens. With the rapid development in the availability of data and computational technologies, various machine learning techniques have been proposed for predicting air pollution. Air Quality Index can be predicted using both Classification and Regression models. Decision tree, Linear Regression, Support Vector Regression and Random Forest Techniques are implemented in this proposed research work. From the above models, it is evident that, Random Forest Regression based model out performs better in the air quality index prediction with an accuracy of 98%.