Comparing performance of Backpropagation and RBF neural network models for predicting daily network traffic

Abstract

The predicting daily network traffic usage is a very important issue in the service activities of the university. This paper present techniques based on the development of backpropagation (BP) and radial basis function (RBF) neural network models, for modelling and predicting the daily network traffic at Universitas Mulawarman, East Kalimantan, Indonesia. The experiment results indicate that a strong agreement between model predictions and observed values, since MSE is below 0.005. When performance indices are compared, the RBFNN-based model is a more accurate predictor with MSE value is 0.00407999, MAPE is 0.03701870, and MAD is 0.06885187 than the BPNN model. Therefore, the smallest MSE value indicates a good method for accuracy, while RBF finding illustrates proposed best model to analyze daily network traffic.