中国安全科学学报 ›› 2021, Vol. 31 ›› Issue (9): 150-156.doi: 10.16265/j.cnki.issn1003-3033.2021.09.021

• 公共安全 • 上一篇    下一篇

基于文本挖掘的重特大交通事故成因网络分析*

韩天园, 田顺** 讲师, 吕凯光, 李旋, 张佳涛, 魏朗 教授   

  1. 长安大学 汽车学院,陕西 西安 710064
  • 收稿日期:2021-06-18 修回日期:2021-08-08 出版日期:2021-09-28 发布日期:2022-03-28
  • 通讯作者: ** 田 顺(1989—),男,江苏灌云人,博士,讲师,主要从事车辆安全工程方面的研究。E-mail:steventian2008@126.com。
  • 作者简介:韩天园 (1996—),男,陕西延安人,硕士研究生,主要研究方向为人-车-路系统安全、道路交通事故分析与预防。E-mail:Hty_vehicle@163.com。
  • 基金资助:
    中央高校基本科研业务费专项(300102220106)。

Network analysis on causes for serious traffic accidents based on text mining

HAN Tianyuan, TIAN Shun, LYU Kaiguang, LI Xuan, ZHANG Jiatao, WEI Lang   

  1. School of Vehicle, Chang'an University, Xi'an Shaanxi 710064, China
  • Received:2021-06-18 Revised:2021-08-08 Online:2021-09-28 Published:2022-03-28

摘要: 为查明重特大交通事故发生的特征,有效预防事故和缓减伤害,通过文本挖掘技术完成对254份重特大道路交通事故调查报告的数据结构化处理;运用改进词频-逆文件频率(TF-IDF)算法,提取出32个高权重关键词;基于事故成因的社会网络中心性分析、核心边缘结构分析和凝聚子群分析,并结合人-车-路系统安全理论,构建重特大道路交通事故机制的层级模型。研究结果表明:重特大道路交通事故成因网络贡献值由大到小依次为违法行为、安全隐患和操作不当,其中违法行为贡献值依次为超载、超速、车道使用错误、疲劳驾驶等;违法行为与操作不当的直接原因和安全隐患的间接原因的耦合,是重特大事故安全运行系统失稳的根本原因。

关键词: 文本挖掘, 重特大交通事故, 社会网络分析, 事故机制, 层级模型

Abstract: In order to study characteristics of serious traffic accidents so as to effectively prevent them and reduce injuries, 254 investigation reports of such accidents were processed by text mining technology. Then,32 high-weight keywords were extracted by using improved term frequency-inverse document frequency (TF-IDF) algorithm. Finally, a hierarchical model of traffic accident mechanism, based on network centrality analysis, core edge structure analysis and agglomerative subgroup analysis on accident causes, was constructed considering theory of human vehicle road system safety. The results show that the contribution values of serious road traffic accidents network rank as illegal behavior, potential safety hazards and improper operation from top to bottom, and in terms of illegal behavior, it goes in the order of overload, speeding, wrong lane use, and fatigue, etc. The coupling of direct causes for illegal behavior and improper operation with indirect ones for potential safety hazards constitute the fundamental reason for failure of safe operation system in major accidents.

Key words: text mining, serious traffic accidents, social network analysis, accident mechanism, hierarchical mode

中图分类号: