面向间断流行程时间预测的浮动车数据挖掘

间断流行程时间预测是交通流诱导系统和交通控制系统研究的一项重要内容。指出传统浮动车行程时间预测模型的局限性，提出一个模糊回归模型，该模型克服了传统预测模型的局限性，考虑了相邻路段交通状态（行程时间）的连续性，仅需要少量数据就可以对间断流行程时间进行较准确的预测。利用杭州市的实测数据对行程时间进行了预测分析，结果证明该模型是有效的。

Urban travel time prediction model based on floating car data mining

Link travel time prediction for the interrupted flow is an essential subject for traffic flow guidance sys- tem and traffic control system. After pointing out the shortcomings of traditional link Travel Time Prediction （TTP） models based on floating car data, the paper brings forward an interrupted flow TTP model based on fuzzy regres- sion. The new model can overcome the drawbacks of traditional models and take traffic state （or link travel time） continuity between two adjacent links into consideration. At the same time, it only requires a small amount of data to implement the model. The paper utilizes the field data from Hangzhou to validate the proposed model, and the re- sults prove to be quite satisfactory.