数据挖掘算法在交通状态量化及识别的应用

在智能交通系统（ITS）环境下，以交通检测器采集的海量交通流信息为对象，通过数据挖掘技术即数据获取、数据预处理、挖掘方法、结果分析与评价、模式应用等进行新的信息提取，提出了各阶段的要求和聚类分析及模式识别的算法，最后从海量数据中得到新的有用信息交通状态分类，同时使用实时采集交通流数据进行交通状态判别。实验结果表明识别状态能够准确反映实际交通状态。

Application of data mining in traffic state quantification and recognition

In the Intelligent Transportation System (ITS) environment, the magnanimous flow information of traffic detector was targeted, using data mining technology which included data collection, data preprocessing, data mining, result analysis and appraisal and pattern application to carry on new information extraction. Various stages request as well as the clustering analysis and the pattern recognition algorithm were proposed. Finally the new useful information traffic state was obtained from the magnanimous data. Simultaneously the real-time gathering traffic data was used to distinguish the traffic state, which finally indicated the recognition state could accurately reflect the actual traffic state.