|本期目录/Table of Contents|

[1]江耀森,杨超宇,刘晓蕾.基于ChatGLM3-6B的煤矿事故知识图谱构建及应用研究*[J].中国安全生产科学技术,2024,20(10):12-21.[doi:10.11731/j.issn.1673-193x.2024.10.002]
　JIANG Yaosen,YANG Chaoyu,LIU Xiaolei.Research on construction and application of knowledge graph for coal mine accidents based on ChatGLM3-6B[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2024,20(10):12-21.[doi:10.11731/j.issn.1673-193x.2024.10.002]
点击复制

基于ChatGLM3-6B的煤矿事故知识图谱构建及应用研究*
分享到：

《中国安全生产科学技术》[ISSN:1673-193X/CN:11-5335/TB]

卷:
20
期数:
2024年10期
页码:
12-21
栏目:
学术论著
出版日期:
2024-10-30

文章信息/Info

Title:
Research on construction and application of knowledge graph for coal mine accidents based on ChatGLM3-6B
文章编号:
1673-193X(2024)-10-0012-10
作者:
江耀森; 杨超宇; 刘晓蕾
（1.安徽理工大学 人工智能学院，安徽 淮南 232001；
2.安徽理工大学 经济与管理学院，安徽 淮南 232001）
Author(s):
JIANG Yaosen;  YANG Chaoyu;  LIU Xiaolei
(1.School of Artificial Intelligence,Anhui University of Science and Technology,Huainan Anhui 232001,China;
2.School of Economics and Management,Anhui University of Science and Technology,Huainan Anhui 232001,China)
关键词:
知识图谱; ChatGLM3-6B; 煤矿事故; 智能查询; 信息抽取
Keywords:
knowledge graph;  ChatGLM3-6B;  coal mine accident;  intelligent query;  information extraction
分类号:
X936
DOI:
10.11731/j.issn.1673-193x.2024.10.002
文献标志码:
A
摘要:
为推动煤矿领域知识图谱构建及应用向自动化、信息化、智能化方向发展，解决传统方法存在大量标注数据、费时费力、识别效果低等问题，提出基于ChatGLM3-6B的煤矿事故知识图谱的构建方法。通过构建提示模板，使用ChatGLM3-6B模型的上下文学习能力，进行知识抽取以及Neo4j图数据库存储。采用LangChain框架，实现针对煤矿事故知识图谱的智能查询问答。研究结果表明：ChatGLM3-6B在单样本提示和少样本提示学习策略下信息抽取效果都要高于Lattice LSTM，Bert-CRF和RoBERTa-CRF模型，在F1值上分别达到0.801 0，0.853 7，其次在知识问答模块，通过煤矿事故问答案例实验测评发现，达到预期应用效果。研究结果不仅可为煤矿安全领域提供1种高效智能的事故知识图谱构建与应用技术，还可为煤矿管理者在事故防控方面的安全决策提供辅助支持。
Abstract:
To promote the development in the construction and application of knowledge graph in the field of coal mine towards the directions of automation,informatization,and intelligence,and solve the problems of large amount of annotated data,laborious and low recognition effect in traditional methods,a construction method of knowledge graph for coal mine accidents based on ChatGLM3-6B was proposed.By constructing the prompt templates and utilizing the context learning ability of ChatGLM3-6B model,the knowledge extraction was performed,followed by the Neo4j graph database storage.With the help of LangChain framework,the intelligent query question and answer of knowledge graph for coal mine accidents was realized.The results show that the information extraction effect of ChatGLM3-6B under One-Shot Prompt and Few-Shot Prompt learning strategies is higher than that of Lattice LSTM,Bert-CRF and RoBERTa-CRF models,and the F1 values are 0.801 0 and 0.853 7 respectively.Furthermore,in the knowledge question and answer module,it is found that the expected application effect is achieved through the experimental evaluation of coal mine accident question and answer cases.The research results not only provide an efficient and intelligent technology for the construction and application of accident knowledge graphs in the field of coal mine safety,but also offer auxiliary support for safety decision-making by coal mine managers in accident prevention and control.

参考文献/References:

［1］李全生.煤炭生态型露天开采理论与技术体系及其应用［J］.煤炭学报，2024，49(5):2426-2444. LI Quansheng.Theory and technical system of coal ecological openpit mining and its application［J］.Journal of China Coal Society，2024，49(5):2426-2444.
［2］程磊，孙洁.2016—2022年我国煤矿事故统计与规律分析［J］.煤炭工程，2023，55(11):125-129. CHENG Lei，SUN Jie.Statistics and law analysis of coal mine acidents in China from 2016 to 2022［J］.Coal Engineering，2023，55(11):125-129.
［3］李泽荃,刘飞翔,赵嘉良,等.面向煤矿安全隐患文本的预训练语言模型构建［J/OL］.矿业安全与环保,1-10［2024-10-11］.http://kns.cnki.net/kcms/detail/50.1062.TD.20240401.1656.002.html.
［4］王忠强，宋建鑫，余数三，等.基于依存句法分析的智慧矿山知识图谱构建方法［J］.矿业研究与开发，2023，43(10):232-240. WANG Zhongqiang，SONG Jianxin，YU Shusan，et al.A method of constructing knowledge graph of intelligent mines based on dependency syntax analysis［J］.Mining Research and Development，2023，43(10):232-240.
［5］刘鹏，叶帅，舒雅，等.煤矿安全知识图谱构建及智能查询方法研究［J］.中文信息学报，2020，34(11):49-59. LIU Peng，YE Shuai，SU Ya，et al.Coal mine safety:knowledge graph construction and its QA approach［J］.Journal of Chinese Information Processing，2020，34(11):49-59.
［6］付燕，刘致豪，叶鸥.基于煤矿井下不安全行为知识图谱构建方法［J］.工矿自动化，2024，50(1):88-95. FU Yan，LIU Zhihao，YE Ou.A method for constructing aknowledge graph of unsafe behaviors in coal mines［J］.Journal of Mine Automation，2024，50(1):88-95.
［7］ZHAO W X，ZHOU K，LI J，et al.A survey of large language models［EB/OL］.(2023-11-24)［2024-05-20］.https://doi.org/10.48550/arXiv.2303.18223.
［8］WEI X，CUI X，CHENG N，et al.Zero-shot information extraction via chatting with chatgpt［EB/OL］.(2023-02-20)［2024-05-20］.https://doi.org/10.48550/arXiv.2302.10205.
［9］DONG Q，LI L，DAI D，et al.A survey on in-context learning［EB/OL］.(2023-06-01)［2024-05-20］.https://doi.org/10.48550/arXiv.2301.00234.
［10］刘峤，李杨，段宏，等.知识图谱构建技术综述［J］.计算机研究与发展，2016，53(3):582-600. LIU Qiao，LI Yang，DUAN Hong，et al.Knowledge graph construction techniques［J］.Journal of Computer Research and Develo-pment，2016，53(3):582-600.
［11］阮彤，王梦婕，王昊奋，等.垂直知识图谱的构建与应用研究［J］.知识管理论坛，2016，1(3):226-234. RUAN Tong，WANG Mengjie，WANG Haofen，et al.Research on the construction and application of vertical knowledge graphs ［J］.Knowledge Management Forum，2016，1(3):582-600.
［12］郑晓芳，杨娜，傅军栋.面向电气照明规范的知识图谱构建［J］.计算机应用与软件，2024，41(2):117-122，129. ZHENG Xiaofang，YANG Na，FU Jundong，et al.Construction of knowledge graph for electrical lighting code［J］.Computer Applications and Software，2024，41(2):117-122，129.
［13］VASWANI A，SHAZEER N，PARMAR N，et al.Attention is all you need［EB/OL］.(2017-06-12)［2024-05-20］.https://doi.org/10.48550/arXiv.1706.03762.
［14］PAN S，LUO L，WANG Y，et al.Unifying large language models and knowledge graphs:a roadmap［EB/OL］.(2023-06-14)［2024-05-20］.https://doi.org/10.48550/arXiv.2306.08302.
［15］PAN J Z，RAZNIEWSKI S，KALO J C，et al.Large language models and knowledge graphs:opportunities and challenges［EB/OL］.(2023-08-11)［2024-05-20］.https://doi.org/10.48550/arXiv.2308.06374.
［16］国家矿山安全监察局.国家矿山安全监察局关于印发《矿山生产安全事故报告和调查处理办法》的通知［EB/OL］.(2023-01-17)［2024-05-20］.https://www.chinamine-safety.gov.cn/zfxxgk/fdzdgknr/tzgg/202301/t20230118_440874.shtml.
［17］LIU P，YUAN W，FU J，et al.Pre-train，prompt，and predict:a systematic survey of prompting methods in natural language processing［EB/OL］.(2021-07-28)［2024-05-20］.https://doi.org/10.48550/arXiv.2107.13586.
［18］国家矿山安全监察局.陕西延安延川县新泰煤矿“8·21”重大瓦斯爆炸事故案例［EB/OL］.(2024-04-15)［2024-05-20］.https://www.chinamine-safety.gov.cn/zfxxgk/fdzdgknr/sgcc/sgalks/202404/t20240415_484888.shtml.
［19］ZHANG Y，YANG J.Chinese NER using lattice LSTM［EB/OL］.(2018-05-05)［2024-05-20］.https://doi.org/10.48550/arXiv.1805.02023.

相似文献/References:

[1]姚德志,宋守信.基于文献计量与共词分析法的企业安全文化研究述评[J].中国安全生产科学技术,2017,13(7):180.[doi:10.11731/j.issn.1673-193x.2017.07.029]
　YAO Dezhi,SONG Shouxin.Review on study of safety culture in enterprise based on bibliometric analysis and co-word analysis method[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2017,13(10):180.[doi:10.11731/j.issn.1673-193x.2017.07.029]
[2]王君玲,盛玲玉.中国煤矿企业应急管理研究回溯与前瞻——基于CiteSpace知识图谱的文献计量分析[J].中国安全生产科学技术,2018,14(9):175.[doi:10.11731/j.issn.1673-193x.2018.09.028]
　WANG Junling,SHENG Lingyu.Backtracking and prospect on emergency management research of coal mine enterprises in China—Bibliometric analysis based on CiteSpace knowledge graph[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2018,14(10):175.[doi:10.11731/j.issn.1673-193x.2018.09.028]
[3]梁本部,谢科范.安全疏散中ICT技术应用研究脉络与热点分析[J].中国安全生产科学技术,2019,15(9):50.[doi:10.11731/j.issn.1673-193x.2019.09.008]
　LIANG Benbu,XIE Kefan.Research context and hotspots analysis of ICT application in safety evacuation[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2019,15(10):50.[doi:10.11731/j.issn.1673-193x.2019.09.008]
[4]卢颖,路越茗,余易凡,等.踩踏事故预防监控领域科学知识图谱研究*[J].中国安全生产科学技术,2021,17(7):172.[doi:10.11731/j.issn.1673-193x.2021.07.028]
　LU Ying,LU Yueming,YU Yifan,et al.Research on scientific knowledge map in field of prevention and monitoring on stampede accidents[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2021,17(10):172.[doi:10.11731/j.issn.1673-193x.2021.07.028]
[5]韩国元,冷雪忠.国内公共卫生安全研究的文献计量分析*[J].中国安全生产科学技术,2022,18(1):17.[doi:10.11731/j.issn.1673-193x.2022.01.003]
　HAN Guoyuan,LENG Xuezhong.Bibliometric analysis of domestic public health security research[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2022,18(10):17.[doi:10.11731/j.issn.1673-193x.2022.01.003]
[6]张玉明,唐艳娟,杜海波,等.Near-miss在知识图谱下的演化路径与发展前沿可视化分析[J].中国安全生产科学技术,2022,18(7):219.[doi:10.11731/j.issn.1673-193x.2022.07.032]
　ZHANG Yuming,TANG Yanjuan,DU Haibo,et al.Visual analysis on evolution path and development frontiers of Near-miss under knowledge map[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2022,18(10):219.[doi:10.11731/j.issn.1673-193x.2022.07.032]
[7]李聪,徐子烜,王雨情,等.城市燃气管网泄漏事故分析知识图谱构建及应用研究*[J].中国安全生产科学技术,2022,18(10):5.[doi:10.11731/j.issn.1673-193x.2022.10.001]
　LI Cong,XU Zixuan,WANG Yuqing,et al.Construction and application of knowledge graph for leakage accident analysis of urban gas pipeline network[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2022,18(10):5.[doi:10.11731/j.issn.1673-193x.2022.10.001]
[8]周义棋,刘畅,龙增,等.电网应急预案知识图谱构建方法与应用*[J].中国安全生产科学技术,2023,19(1):5.[doi:10.11731/j.issn.1673-193x.2023.01.001]
　ZHOU Yiqi,LIU Chang,LONG Zeng,et al.Construction method and application of knowledge graph in emergency plans for power grid[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2023,19(10):5.[doi:10.11731/j.issn.1673-193x.2023.01.001]
[9]许慧,廖桧铭.基于知识图谱的建筑工程施工事故智能分析*[J].中国安全生产科学技术,2023,19(3):46.[doi:10.11731/j.issn.1673-193x.2023.03.007]
　XU Hui,LIAO Huiming.Intelligent analysis of construction accidents in construction engineering based on knowledge graph[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2023,19(10):46.[doi:10.11731/j.issn.1673-193x.2023.03.007]
[10]徐方廷,黄锐,刘华森,等.21世纪矿山职业安全健康研究科学知识图谱*——演进历程、热点前沿及发展态势（2001—2021年）[J].中国安全生产科学技术,2023,19(6):52.[doi:10.11731/j.issn.1673-193x.2023.06.008]
　XU Fangting,HUANG Rui,LIU Huasen,et al.Scientific knowledge map of mine occupational safety and health research in 21st century:evolution process,hot front and development trend (2001—2021)[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2023,19(10):52.[doi:10.11731/j.issn.1673-193x.2023.06.008]

备注/Memo

备注/Memo:
收稿日期： 2024-05-20
* 基金项目： 国家自然科学基金项目(61873004)
作者简介： 江耀森，硕士研究生，主要研究方向为煤矿安全智能化建设、大语言模型。
通信作者： 杨超宇，博士，教授，主要研究方向为大数据分析与挖掘、煤矿安全风险管理。
更新日期/Last Update: 2024-10-31