基于规则的城市轨道交通安全事件信息抽取及其知识元表示  

Rule-based Information Extraction of Urban Rail Transit Safety Cases and Its Common Knowledge Meta-model Representation

作　　者：张萌[1,2] 陈佳惠 孙然然 李晓璐 朱广宇[1,2] ZHANG Meng;CHEN Jia-hui;SUN Ran-ran;LI Xiao-lu;ZHU Guang-yu(Beijing Research Center of Urban Traffic Information Sensing and Service Technologies, Beijing Jiaotong University, Beijing 100044, China;Key Laboratory of Transport Industry of Big Data Application Technologies for Comprehensive Transport, Beijing Jiaotong University, Beijing 100044, China)

机构地区：[1]北京交通大学,北京市城市交通信息智能感知与服务工程技术研究中心,北京100044 [2]北京交通大学,综合交通运输大数据应用技术交通运输行业重点实验室,北京100044

出　　处：《科学技术与工程》2021年第15期6435-6440,共6页Science Technology and Engineering

基　　金：国家重点研发计划(2016YFB1200203-02);国家自然科学基金(61833002;61872037);中央高校基本科研业务费专项(2021JBM403)。

摘　　要：合理利用城市轨道交通安全事件案例对突发事件下辅助制定应急决策具有重大意义。目前,中国轨道交通运营商存储了大量的安全事件案例,但大多以自由化或半自由化文本的形式存储在数据库中,使用率较低。为提高城市轨道交通安全事件案例的使用效率,提出了基于规则的信息抽取方法,将城市轨道交通安全事件案例的自由文本转化为用共性知识元表示的结构化数据。通过指标计算,所提出的方法可以以较高的准确率和召回率对安全事件要素进行抽取,可为突发事故时的应急决策制定提供高效的数据支撑。It is of great significance to assist in emergency decision-making by making use of urban rail transit safety event cases reasonably.At present,a large number of security cases are stored by China's rail transit operators store,but most of them are stored in the database in the form of free or semi-free texts,with low utilization rate.In order to improve the use efficiency of urban rail transit safety cases,a rule-based information extraction method was proposed.It converted the free texts of urban rail transit safety cases into structured data represented by common knowledge elements.Through the index calculation,the elements of safety events can be extracted by the proposed method with high accuracy and recall rate.And efficient data support for emergency decision-making can be provided.

关 键 词：城市轨道交通安全事件案例 共性知识元 抽取规则 信息抽取 

分 类 号：U231.92[交通运输工程—道路与铁道工程]