Construction of a mine accident knowledge graph based on Large Language Models

Current methods for constructing knowledge graphs in the field of mining require a large amount of manually labeled high-quality supervised data during the pre-training stage, resulting in high labor costs and low efficiency. Large Language Models (LLMs) can significantly improve the quality and eff...

Full description

Saved in:
Bibliographic Details
Main Authors: ZHANG Pengyang, SHENG Long, WANG Wei, WEI Zhongcheng, ZHAO Jijun
Format: Article
Language:zho
Published: Editorial Department of Industry and Mine Automation 2025-02-01
Series:Gong-kuang zidonghua
Subjects:
Online Access:http://www.gkzdh.cn/article/doi/10.13272/j.issn.1671-251x.2024080031
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849311203619241984
author ZHANG Pengyang
SHENG Long
WANG Wei
WEI Zhongcheng
ZHAO Jijun
author_facet ZHANG Pengyang
SHENG Long
WANG Wei
WEI Zhongcheng
ZHAO Jijun
author_sort ZHANG Pengyang
collection DOAJ
description Current methods for constructing knowledge graphs in the field of mining require a large amount of manually labeled high-quality supervised data during the pre-training stage, resulting in high labor costs and low efficiency. Large Language Models (LLMs) can significantly improve the quality and efficiency of information extraction with only a small amount of manually labeled high-quality data. However, the prompt-based approach in LLMs suffers from catastrophic forgetting. To address this issue, graph-structured information was embedded into the prompt template and a Graph-Structured Prompt was proposed. By integrating this prompt into the LLM, high-quality construction of a mine accident knowledge graph based on the LLM was achieved. First, publicly available mine accident reports were collected from the Coal Mine Safety Production Network and preprocessed through formatting corrections and redundant information removal. Next, the LLM was utilized to extract knowledge embedded in the accident reports and K-means clustering was used to classify entities and relationships, thereby completing the construction of the mine accident ontology. Then, a small amount of data were labeled based on the ontology, which was used for LLM training and fine-tuning. Finally, the LLM embedded with the Graph-Structured Prompt was employed for information extraction, instantiating entity-relation triples to construct the mine accident knowledge graph. Experimental results showed that LLMs outperformed the Universal Information Extraction (UIE) model in entity and relationship extraction tasks. Moreover, the LLM embedded with the Graph-Structured Prompt achieved higher precision, recall, and F1 scores compared to those without it.
format Article
id doaj-art-d51a4469ff1f4cf2bec49b739ccecbb7
institution Kabale University
issn 1671-251X
language zho
publishDate 2025-02-01
publisher Editorial Department of Industry and Mine Automation
record_format Article
series Gong-kuang zidonghua
spelling doaj-art-d51a4469ff1f4cf2bec49b739ccecbb72025-08-20T03:53:28ZzhoEditorial Department of Industry and Mine AutomationGong-kuang zidonghua1671-251X2025-02-015127683, 10510.13272/j.issn.1671-251x.2024080031Construction of a mine accident knowledge graph based on Large Language ModelsZHANG PengyangSHENG LongWANG WeiWEI ZhongchengZHAO JijunCurrent methods for constructing knowledge graphs in the field of mining require a large amount of manually labeled high-quality supervised data during the pre-training stage, resulting in high labor costs and low efficiency. Large Language Models (LLMs) can significantly improve the quality and efficiency of information extraction with only a small amount of manually labeled high-quality data. However, the prompt-based approach in LLMs suffers from catastrophic forgetting. To address this issue, graph-structured information was embedded into the prompt template and a Graph-Structured Prompt was proposed. By integrating this prompt into the LLM, high-quality construction of a mine accident knowledge graph based on the LLM was achieved. First, publicly available mine accident reports were collected from the Coal Mine Safety Production Network and preprocessed through formatting corrections and redundant information removal. Next, the LLM was utilized to extract knowledge embedded in the accident reports and K-means clustering was used to classify entities and relationships, thereby completing the construction of the mine accident ontology. Then, a small amount of data were labeled based on the ontology, which was used for LLM training and fine-tuning. Finally, the LLM embedded with the Graph-Structured Prompt was employed for information extraction, instantiating entity-relation triples to construct the mine accident knowledge graph. Experimental results showed that LLMs outperformed the Universal Information Extraction (UIE) model in entity and relationship extraction tasks. Moreover, the LLM embedded with the Graph-Structured Prompt achieved higher precision, recall, and F1 scores compared to those without it.http://www.gkzdh.cn/article/doi/10.13272/j.issn.1671-251x.2024080031mine accidentknowledge graphlarge language modelgraph-structured promptontology constructioninformation extraction
spellingShingle ZHANG Pengyang
SHENG Long
WANG Wei
WEI Zhongcheng
ZHAO Jijun
Construction of a mine accident knowledge graph based on Large Language Models
Gong-kuang zidonghua
mine accident
knowledge graph
large language model
graph-structured prompt
ontology construction
information extraction
title Construction of a mine accident knowledge graph based on Large Language Models
title_full Construction of a mine accident knowledge graph based on Large Language Models
title_fullStr Construction of a mine accident knowledge graph based on Large Language Models
title_full_unstemmed Construction of a mine accident knowledge graph based on Large Language Models
title_short Construction of a mine accident knowledge graph based on Large Language Models
title_sort construction of a mine accident knowledge graph based on large language models
topic mine accident
knowledge graph
large language model
graph-structured prompt
ontology construction
information extraction
url http://www.gkzdh.cn/article/doi/10.13272/j.issn.1671-251x.2024080031
work_keys_str_mv AT zhangpengyang constructionofamineaccidentknowledgegraphbasedonlargelanguagemodels
AT shenglong constructionofamineaccidentknowledgegraphbasedonlargelanguagemodels
AT wangwei constructionofamineaccidentknowledgegraphbasedonlargelanguagemodels
AT weizhongcheng constructionofamineaccidentknowledgegraphbasedonlargelanguagemodels
AT zhaojijun constructionofamineaccidentknowledgegraphbasedonlargelanguagemodels