A Dynamic Traffic Forecast Using Hybrid Wavelet Network with an Adaptive Genetic Local Search

The paper proposes a dynamic predictor using hybrid wavelet neural network, which use multiple types of mother wavelet in the network according to the feature of pattern change of predicted information instead of a single type. The predictor has function of parallel prediction. In addition, an adaptive genetic local search algorithm is employed to train effectively the network. A case study of traffic information forecasting using proposed wavelet neural network is demonstrated.