基于神经网络的交通信息融合预测方法

为了更加准确地预测动态变化的交通信息，通过分析城市道路交通流量变化的特点，提出一种基于神经网络的融合预测方法。这种方法根据预测数据各属性的特点，将数据构造为多个相关的时间序列。在此基础上，对各序列采取不同的处理方法，然后利用神经网络进行融合，得到最终的预测结果。这种方法可用于数据动态预测的各种领域。实验结果表明，采用这种方法可以有效地改善数据预测的误差。

Data-Fusion Prediction of Traffic Information Based on Artificial Neural Network

In this paper, an improved method of dynamic data prediction, with data fusion based on artificial neural network, is proposed. It obtains a more precise prediction of the varying traffic information. Multiple correlative serials distilled from the collected data are treated with respective methods, from which the end result is predicted with data fusion on the basis of artificial neural network. The accurate prediction is also very appealing in other fields of dynamic data prediction. Experimental results have shown that the method is able to evidently decrease errors.