基于知识图谱的煤矿风险管控人员推荐算法研究  

Research on Recommendation Algorithm of Coal Mine Risk Management and Control Personnel Based on Knowledge Mapping

作　　者：赵红泽[1,2] 郑雯 郝强 郭卫洪 顾书豪 马新根 ZHAO Hongze;ZHENG Wen;HAO Qiang;GUO Weihong;GU Shuhao;MA Xingen(School of Energy and Mining Engineering,China University of Mining and Technology-Beijing,Beijing 100083,China;State Key Laboratory for Geomechanics and Deep Underground Engineering,China University of Mining and TechnologyBeijing,Beijing 100083,China;Huaneng Coal Technology Research Co.,Ltd.,Beijing 100070,China)

机构地区：[1]中国矿业大学(北京)能源与矿业学院,北京100083 [2]中国矿业大学(北京)深部岩石力学与地下工程国家重点实验室,北京100083 [3]华能煤炭技术研究有限公司,北京100070

出　　处：《煤炭技术》2024年第11期202-207,共6页Coal Technology

基　　金：国家重点研发计划资助(2022YFB4703701);中国矿业大学(北京)深部岩土力学与地下工程国家重点实验室开放基金课题资助(SKLGDUEK1923)。

摘　　要：随着智能化、信息化煤矿安全管控系统的普遍应用,迅速准确将风险分门别类地推荐给相应煤矿风险管控人员成为安全管控的问题之一。提出了一种基于知识图谱的煤矿风险管控人员推荐算法。首先对煤矿现有风险管控人员与历史风险信息属性进行研究,建立风险知识图谱,获取风险管控人员-风险信息匹配度矩阵;然后,利用用户的基本特征与知识图谱提取特征进行融合,绘制用户画像;以用户画像为主体,结合协同过滤算法进行混合推荐;基于数据集设计实施对照实验,证明该算法在准确率、召回率、误差平均绝对值上的优越性。实验结果表明,算法能够有效地在多个维度中进行推荐,实现数据的充分利用和推荐多样性,有效提升了推荐性能。With the widespread application of intelligent and information-based coal mine safety management and control system,it has become one of the problems of safety management and control to quickly and accurately recommend risks to corresponding coal mine risk management and control personnel.Propose a recommendation algorithm for coal mine risk management and control personnel based on knowledge map.Firstly,the existing risk management and control personnel and historical risk information attributes of the coal mine are studied,the risk knowledge map is established,and the risk management and control personnel risk information matching matrix is obtained;Then,the user's basic features and knowledge maps are used to extract features for fusion and draw user portraits;The user profile is taken as the main body,combined with collaborative filtering algorithm for hybrid recommendation;Based on the data set design and implementation of the contrast experiment,it is proved that the algorithm is superior in accuracy,recall,and the average absolute value of error.The experimental results show that the algorithm can effectively recommend in multiple dimensions,achieve full utilization of data and diversity of recommendations,and effectively improve the recommendation performance.

关 键 词：煤矿 知识图谱 推荐算法 协同过滤 风险管控 

分 类 号：TD679[矿业工程—矿山机电]