Prediction of concentration of SOx, NOx and PM10 due to vehicular emission using CART (Regression Tree) analysis

Urbanization and industrial expansion are two factors which contribute to an enormous rise in air pollution. Even mountainous towns like Ranchi in Jharkhand, India have seen a decline in air quality. Urbanization leads to an increase in traffic density. The primary pollutants which contribute to air pollution from vehicular emissions are SO X , NO X & PM 10 . An attempt is made in this research to estimate the concentration of the above pollutants using regression tree analysis (CART). Statistical parameters like Index of agreement (d), Net Absolute Error (NAE), Mean Absolute Error (MAE), Mean Square Error (MSE), Root Mean Square Error (RMSE) and Coefficient of Determination (R 2 ) are used to validate the accuracy of models.