关键词: 交通运输系统工程, 交通流量预测, 统计学习理论, 支持向量机, BP神经网络

Abstract: Summarizing the existing prediction models and considering the characteristics of the traffic itself such as nonlinearity, complexity and uncertainty, a shortterm traffic flow prediction model based on support vector machine(SVM) was proposed. The real traffic data validation and the comparison with the traffic flow prediction model based on the BP neural network resulted in a conclusion that this model is superior to the BP neural network model on the aspects of prediction accuracy, convergence time, generalization ability, optimization possibility and so on.

Key words: engineering of communications and transportation system, traffic flow prediction, statistics learning theory, SVM, BP neural network