Published Paper Details:

The rapid development in commercial and social areas of modern society has affected the air quality adversely. The pollutant concentration due to industries and transport continue to grow and have an effect on human life. According to the global pollution database maintained by the World Health Organization (WHO), India has the greatest number of polluted cities globally. Higher levels of the Air Quality Index (AQI), which indicates increased risk due to pollutant exposure, are an important indicator that directly affects public health. Thus, tracking air quality index is essential in developing countries so that measures to curb air pollution could be tackled. In this research paper the primary aim is to anticipate the levels of Air Quality Index using classification techniques namely Decision Tree, Random Forest, Linear Regression and Support Vector Machines. For the experimental work the dataset of Delhi: Ashok Vihar and 2 stations of Kolkata: Ballygunge and Victoria has been gathered from Central Pollution Control Board Website. It has been concluded that Random Forest depicts maximum performance than the other classification techniques based on accuracy, precision and recall and SVM exhibits the lowest prediction accuracy.