Concentration Separation Prediction Model to Enhance Prediction Accuracy of Particulate Matter

Demand for more accurate particulate matter forecasts is accumulating owing to the increased interest and issues regarding matter. Incredibly low concentration matter, which accounts most of overall often underestimated when a prediction model based on machine learning used. This study proposed concentration-specific separation overcome this shortcoming. Three models Deep Neural Network (DNN), Recurrent (RNN), Long Short-Term Memory (LSTM), commonly used performance evaluation model, were as comparative models. Root mean squared error (RMSE), absolute percentage (MAPE), accuracy utilized evaluation. The results showed that all Air Quality Index (AQI) segments was than 80 percent in entire spectrum. In addition, confirmed over-prediction phenomenon single neural network concentrated ‘normal’ AQI region alleviated.