基于云-自组织神经网络的交通流预测模型

现代交通系统结构复杂，涉及的数据类型和数量众多，模糊性、随机性和不确 定性等因素的存在增加了数据分析过程中定性与定量综合集成的难度.本文对城市交通 流预测进行了研究，根据云模型和自组织神经网络的特点，构建了云-自组织神经网络 交通流预测模型.该预测模型运用云模型处理数据的模糊性和随机性问题的优势，提高了 自组织神经网络预测中学习样本数据的可靠性.通过对某城区的实际数据进行对比测算， 改进的预测模型比单纯使用自组织神经网络预测模型决定系数更高.结果表明，本文提出 的模型在交通流预测中提高了准确率，降低了预测泛化误差.

Traffic Flow Forecasting Model Based on Cloud-Self-Organizing Neural Network

Modern transportation systems have complex structure, and the existence of fuzzy, stochastic and uncertainty factors increase the difficulty of huge data involved in qualitative and quantitative integrated analysis. This paper developed the cloud neural network self-organization of traffic flow forecasting model based on the characteristics of cloud model and self-organizing neural network. Using cloud model fuzziness and randomness advantages, the paper proposed the prediction model that can improve the reliability of selforganizing neural network prediction learning sample data to process data problems. Through comparing two models to a city traffic flow forecasting with actual data, the paper found that the forecasting model has higher coefficient of determination than the only using of self-organizing neural network. The results show that the model proposed in the traffic flow forecasting can improve accuracy and reduce generalization error.