One Step Ahead Prediction of Ozone Concentration for Determination of Outdoor Air Quality Level

With the rapid spread of urbanization, competent authorities become increasingly anxious from air pollution risks and effect on citizens especially those with respiratory diseases. In this work, performances six machine learning methods were analyzed for prediction maximum ozone (O_3) concentration next-day. The models make using concentrations atmospheric components (PM2.5, PM10, Ozone (O3), Sulfur Dioxide (SO2), Nitrogen (NO2), Carbon Monoxide (CO)). utilized are multilayer perception (MLP), Support Vector Regression (SVM), k-Nearest Neighbor (K-NN), Random Forests (RF), Gradient Boosting (GB), Elastic Net (EN). After predictions made by these models, predicted values further processed to be classified into one quality levels defined United States Environmental Protection Agency. as well their corresponding classification results analyzed. It was shown that MLP model gives lowest RMSE 2246 step while SVR achieved highest accuracy score 0.790.