Using the Hybrid MLR-RNN Approach for Air Pollution Forecasting

Air Quality Modeling gained great importance in atmospheric pollution because of its negative effects on the environment and human health. In our study, the relationship between (Particulate Matter PM10) and other nine variables over three years is studied to applied the multiple linear regression models (MLR). The MLR model is the most common for studying like this multivariate case. The main problem for this type of data is the non linear style that has been referred by many researchers before. The recurrent neural network (RNN) is nonlinear methodwhich can be used to solve the nonlinearity problem and resul better forecasting. The hybrid method MLRRNN can be used also for the best results and lead to more accurate forecasting. The hybrid method MLRRNN has improved the performance of MLR method separately