作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2011, Vol. 37 ›› Issue (20): 191-193. doi: 10.3969/j.issn.1000-3428.2011.20.066

• 人工智能及识别技术 • 上一篇    下一篇

交通流数据清洗规则研究

王晓原,张敬磊,吴 芳   

  1. (山东理工大学交通与车辆工程学院智能交通研究所,山东 淄博 255049)
  • 收稿日期:2011-03-21 出版日期:2011-10-20 发布日期:2011-10-20
  • 作者简介:王晓原(1970-),男,教授、博士,主研方向:交通运输系统仿真、建模及其优化;张敬磊,讲师、硕士;吴 芳,硕士研究生
  • 基金资助:
    国家自然科学基金资助项目(61074140);山东省自然科学基金资助项目(ZR2010FM007)

Research on Traffic Flow Data Cleaning Rules

WANG Xiao-yuan, ZHANG Jing-lei, WU Fang   

  1. (Institute of Intelligent Transportation, School of Transportation and Vehicle Engineering, Shandong University of Technology, Zibo 255049, China)
  • Received:2011-03-21 Online:2011-10-20 Published:2011-10-20

摘要: 交通检测器获得的数据存在无效、冗余、错误、时间点漂移及丢失等质量问题。为此,在分析影响数据质量问题原因的基础上,给出交通流数据清洗的概念,研究“脏数据”的清洗规则与清洗步骤,并对环形线圈检测器检测到的数据进行验证。结果表明,该清洗规则对错误、丢失、冗余等“脏数据”的识别率均在90%以上。

关键词: 交通流, 智能运输系统, 数据质量, 数据清洗, 规则

Abstract: Aiming at that many quality problems are existed inevitably in detected data, including inefficacy, redundancy, error, missing, time dot excursion etc, the definition of data cleaning is proposed on the basis of sufficient study and analysis for the influence reasons of data quality, the cleaning rules and cleaning steps of “dirty data” are studied at the same time. And the proposed cleaning rules are calibrated with the detected data of loop vehicle detector. Results show that the recognition rates of “dirty data” is up to 90%.

Key words: traffic flow, Intelligent Transportation Systems(ITS), data quality, data cleaning, rules

中图分类号: