首页 | 本学科首页   官方微博 | 高级检索  
     

一种基于频繁序列匹配的交通状态趋势预测方法
引用本文:杜瑾,;郝珺,;成菊芳. 一种基于频繁序列匹配的交通状态趋势预测方法[J]. 电子设计工程, 2014, 0(15): 15-18
作者姓名:杜瑾,  郝珺,  成菊芳
作者单位:[1]长安大学信息工程学院,陕西西安710064; [2]西安铁路局,陕西西安710054; [3]陕西安达综合性能检测站,陕西西安710068
基金项目:陕西省自然科学基金($2009JCl039);中央高校基本科研业务费专项资金(CHD2011JC021)
摘    要:海量的交通流数据中一定隐藏着某些潜在的交通状态演变规律,然而少有研究能以实验的方法支持这种观点。本文提出一种基于频繁序列匹配的交通状态趋势预测方法:首先,介绍交通状态序列模型及序列切分处理,其次,提出投影压缩序列相似度计算方法及序列匹配算法,第三,讨论基于频繁序列匹配的交通序列预测算法。通过真实采集数据验证本文提出的方法可行有效。

关 键 词:交通流  状态预测  频繁序列模式  序列相似度

The method of traffic status prediction based on frequent traffic status sequence matching
Affiliation:DU Jin, HAO Jun, CHENG Ju-fang (1. School of Information Engineering, Chang'an University, Xi'an 710064, China; 2. Bureau of Xi' an Rail-way, Xi'an 710054, China; 3. Shaanxi Anda Comprehensive Function Measuring Station, Xi'an 710068, China)
Abstract:As known that some latent laws of traffic status evolvement must exist among huge traffic data. However, little work has been done to support this view by experimental methods. In this paper, a new approach is presented to investigate the traffic flow prediction on the basis of traffic status frequent sequence similarity. Firstly, the concepts of traffic status frequent sequences and the algorithm of sequence similarity were introduced. Secondly, the measure of frequent sequence matching was proposed. Thirdly, the traffic status prediction was discussed on the basis of similarity matching. Lastly, the efficiency and validity of this method was proved by real experiment data observed.
Keywords:traffic flow  status prediction  frequent sequence  sequence similarity
本文献已被 维普 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号