煤矿事故案例存储与检索研究  

Study on storage and retrieval of coal mine accident cases

作　　者：杨梦 陈宁[2] 范誉航 YANG Meng;CHEN Ning;FAN Yuhang(School of Mechanical Electronic and Information Engineering,China University of Mining and Technology-Beijing,Beijing 100083,China;Institute of Remote Sensing and Geographic Information System,Peking University,Beijing 100871,China;International school,Beijing University of Post and Telecommunications,Beijing 100876,China)

机构地区：[1]中国矿业大学(北京)机电与信息工程学院,北京100083 [2]北京大学遥感与地理信息系统研究所,北京100871 [3]北京邮电大学国际学院,北京100876

出　　处：《煤炭科学技术》2021年第9期103-109,共7页Coal Science and Technology

基　　金：国家重点研发计划重点专项资助项目(2016YFC0801800)。

摘　　要：为解决事故案例非结构化、多源异构、难以共享的问题,提高事故案例在应急救援管理中的利用率,利用网络爬虫技术获取由各地监管部门发布在互联网上的大量实时事故案例,通过框架法构建数据结构以表示事故案例蕴含的知识,建立了一个通用、全面、共享的事故案例数据库;在事故案例数据库的基础上,初步提出了一种新的案例检索算法,利用搜索引擎中倒排索引技术实现对案例非结构化数据进行检索,同时结合传统案例相似度计算方式对结构化数据进行匹配,实现利用少量关键信息进行非结构化案例数据的高效筛选,可使系统依据指挥人员意愿结合非结构化数据和结构化数据,进行有侧重、有倾向的案例检索,以中国煤矿安全生产网为例对瓦斯、水灾、火灾事故案例进行自动爬取,实践结果表明,此案例检索流程及算法提高了案例检索的有效性和实用性。In order to solve the problem of unstructured,heterogeneous,and difficult to share accident cases,and to improve the utiliza⁃tion rate of accident cases in emergency rescue management,this paper uses network crawler technology to obtain a large number of realtime accident cases published on the Internet by local regulatory authorities.The framework method constructs a data structure to express the knowledge contained in accident cases,and establishes a universal,comprehensive and shared dynamic database of accident cases;on the basis of accident case database,a new case retrieval algorithm is initially proposed,using the index technology as search engine to re⁃trieve unstructured case data.At the same time,the traditional method of case similarity calculation is used to match structured data,and realize the efficient screening of unstructured case data with a small amount of key information,so that the system can be based on the wi⁃shes of the commander combining unstructured data and structured data,a focused and inclination case search was carried out.Taking China Coal Mine Safety Production Network as an example to automatically crawl gas,flood,and fire accident cases.The practical results show that this case retrieval process and algorithm improve the effectiveness and practicability of case retrieval.

关 键 词：网络爬虫 框架表示法 信息检索 倒排索引 案例相似度计算 

分 类 号：TD77[矿业工程—矿井通风与安全]