关键词: 灰色系统, 神经网络, 支持向量机, 组合预测, 交通量

Abstract: It turns out that the excepted effect gains verification, effectively improves the model accuracy, and makes more exact forecasting, which expects to be helpful to concerned departments and personnel to grasp the traffic market trend or make decisions. The grey system forecasting model, neural network forecasting model and support vector machine forecasting model are proposed in this paper. Taking the road goods traffic volume from the year of 1996 to 2003 in the whole country as a study case, the forecasting results are got by three methods. From the forecasting results, we can conclude that the accuracy of the support vector machine forecasting method is higher. Analyzing the characteristic of the combining forecasting method, based on the grey system forecasting model, neural network forecasting model and support vector machine forecasting model, the linear combining forecasting model and support vector machine combining forecasting model are set up. Compared with single prediction methods and linear combining forecasting method, the accuracy of the support vector machine combining forecasting method is higher.

Key words: grey system, neural network, support vector machine, combining forecasting, traffic volume