combinatorial predictive models;  nearest neighbor clustering algorithm;  Radial basis;  RBF neural networks;  Traffic flow

According to the Radial Basis (RBF) neural network and autoregressive sum sliding average (ARIMA) based on nearest neighbor clustering algorithm (NNCA), their respective single-term prediction submodels are established, and the RBF neural network is used to combine the results of the two single-term prediction submodels to obtain the final prediction value. The experimental results prove the effectiveness of the model by applying the model to the traffic flow prediction of a certain road section in Changsha, and the prediction accuracy of the combined prediction model is greatly improved compared with the single prediction model.