Comparative study among three different artificial neural networks to infectious diarrhea forecasting

Infectious diarrhea is an important public health problem around the world. Meteorological factors have been strongly linked to the incidence of infectious diarrhea. Therefore, accurately forecast infectious diarrhea under the effect of meteorological factors is critical to control efforts. In this paper, the abilities of three different artificial neural network (ANN) models, including feed forward back-propagation neural network (FFBPNN), radial basis function neural network (RBFNN) and generalized regression neural network (GRNN), to predict the daily number of infectious diarrhea belonging to the city of Shanghai in China using meteorological factors as input parameters are investigated. All predicting models were developed and tested using same historical dataset (2005-2008). The meteorological factors include temperature, relative humidity, atmospheric pressure, wind speed, sunshine duration and rainfall. As a comparison, a multiple linear regression (MLR) model is also examined using same dataset. The experimental results demonstrated that the model with the best performance is the FFBPNN, followed by the RBFNN and GRNN. The MLR model is the one with the worst performance. This work shows the advantage of FFBPNN over RBFNN and GRNN for predicting infectious diarrhea. These results also imply that the model-building process should be carefully conducted.