基于大模型与GraphRAG的胶东金矿智能搜索技术

doi:10.13745/j.esf.sf.2025.4.77

地学前缘 ›› 2025, Vol. 32 ›› Issue (4): 155-164.DOI: 10.13745/j.esf.sf.2025.4.77

基于大模型与GraphRAG的胶东金矿智能搜索技术

李博文¹^,²(), 王永志¹^,³^,^*(), 丁正江⁴^,⁵, 王斌⁴^,⁵, 温世博³, 董宇浩¹, 纪政³

1.吉林大学综合信息矿产预测研究所, 吉林长春 130061
2.长春黄金研究院有限公司, 吉林长春 130012
3.吉林大学地球探测科学与技术学院, 吉林长春 130061
4.自然资源部深部金矿勘查开采技术创新中心, 山东威海 264209
5.山东省地质矿产勘查开发局第六地质大队, 山东威海 264209

收稿日期:2025-01-24 修回日期:2025-04-20 出版日期:2025-07-25 发布日期:2025-08-04
通信作者: *王永志(1974—),男,博士,教授,主要从事地球科学大数据分析与挖掘、矿产资源智能预测等理论与应用研究工作。E-mail: wangyongzhi@jlu.edu.cn
作者简介:李博文(2000—),男,硕士研究生,地球探测与信息技术专业,从事数据与AI驱动的矿产资源智能预测研究。E-mail: 630376038@qq.com
基金资助:
国家重点研发计划项目(2023YFC2906903);国家重点研发计划项目(2021YFC2901801);国家重点研发计划项目(2023YFC2907105);国家自然科学基金重点项目(42230810);自然资源部科技支撑项目(ZKKJ202419);山东省地矿局科技攻关项目(KY202502)

Intelligent search technology for Jiaodong gold deposits based on large models and GraphRAG

LI Bowen¹^,²(), WANG Yongzhi¹^,³^,^*(), DING Zhengjiang⁴^,⁵, WANG Bin⁴^,⁵, WEN Shibo³, DONG Yuhao¹, JI Zheng³

1. Integrated Information Mineral Prediction Research Institute, Jilin University, Changchun 130061, China
2. Changchun Gold Research Institute Co.,Ltd., Changchun 130012, China
3. College of Geoexploration Science and Technology, Jilin University, Changchun 130061, China
4. Ministry of Natural Resources Technology Innovation Center for Deep Gold Resources Exploration and Mining, Weihai 264209, China
5. No.6 Geological Team of Shandong Provincial Bureau of Geology and Mineral Resources, Weihai 264209, China

Received:2025-01-24 Revised:2025-04-20 Online:2025-07-25 Published:2025-08-04

摘要/Abstract

摘要：

胶东金矿是我国东部重要的金矿资源集中区,其地质信息复杂、知识体系庞大,传统的信息检索方式难以满足矿产勘查中对语义理解与知识推理的高阶需求。为提升地质知识服务效率,本文基于GraphRAG(知识图谱增强型检索生成)技术,构建了面向胶东金矿领域的智能搜索问答系统。研究以知网上胶东金矿相关的论文为语料来源,利用OCR与大语言模型(LLM)技术进行文本解析与语义标准化处理,形成覆盖矿化类型、控矿构造、矿物组合等核心概念的本体知识体系。系统通过提示工程驱动的大模型实现实体与关系自动抽取,构建结构化知识图谱,并集成于图数据库Neo4j中。进一步融合语义嵌入与社区聚类算法,构建知识索引网络,支持自然语言问答、语义扩展与知识溯源等功能。评估结果表明:该系统在回答准确性、上下文精度与知识可解释性等方面优于传统RAG方法及ChatGPT-4o等通用模型,具备更高的专业适应性和推理能力。研究结果可为金矿领域的智能化信息服务提供新型技术路径,也为图谱增强语言模型在地学知识管理中的应用探索提供理论支持。

关键词: GraphRAG, 知识图谱, 大语言模型, 胶东金矿, 知识问答

Abstract:

The Jiaodong gold deposit is a major concentration area of gold resources in eastern China, characterized by complex geological information and an extensive knowledge system. Traditional information retrieval methods struggle to meet the advanced demands of semantic understanding and knowledge reasoning in mineral exploration. To enhance geological knowledge service efficiency, this study develops an intelligent question-answering system for the Jiaodong gold deposit domain based on GraphRAG (Graph-enhanced Retrieval-Augmented Generation) technology. The research utilizes academic papers from CNKI as the corpus, employs OCR and large language models (LLMs) for text parsing and semantic standardization to establish an ontological knowledge system covering core concepts such as mineralization types, ore-controlling structures, and mineral assemblages. The system uses prompt engineering-driven LLMs to automatically extract entities and relationships, constructing a structured knowledge graph integrated into Neo4j. Furthermore, by combining semantic embedding with community clustering algorithms, a knowledge indexing network enables natural language question answering, semantic query expansion, and knowledge provenance. Evaluation results demonstrate the system’s superiority over traditional RAG and general models (e.g., ChatGPT-4o) in answer accuracy, contextual precision, and knowledge interpretability, exhibiting enhanced professional adaptability and reasoning capabilities. The findings provide a novel technical pathway for intelligent information services in gold deposits and theoretical support for knowledge of graph-enhanced language models in geoscience knowledge management.

Key words: GraphRAG, knowledge graph, large language model, Jiaodong gold deposit, knowledge question answering

中图分类号:

P628
P618.51

李博文, 王永志, 丁正江, 王斌, 温世博, 董宇浩, 纪政. 基于大模型与GraphRAG的胶东金矿智能搜索技术[J]. 地学前缘, 2025, 32(4): 155-164.

LI Bowen, WANG Yongzhi, DING Zhengjiang, WANG Bin, WEN Shibo, DONG Yuhao, JI Zheng. Intelligent search technology for Jiaodong gold deposits based on large models and GraphRAG[J]. Earth Science Frontiers, 2025, 32(4): 155-164.

图/表 9

图1 胶东金矿智能搜索系统架构

Fig.1 Architecture of the intelligent search system for Jiaodong gold deposits

图2 GraphRAG语义检索优化流程图

Fig.2 Flowchart for optimizing semantic retrieval in GraphRAG

图3 胶东金矿知识图谱自动构建效果子图

Fig.3 Automatic construction effect subgraph of Jiaodong gold deposit knowledge graph

图4 知识图谱自动构建效果细节展示图

Fig.4 Knowledge graph auto-construction details visualization

图5 胶东金矿智能搜索和问答系统工作流程

Fig.5 Workflow of the intelligent question-answer system for Jiaodong gold deposits

图6 胶东金矿智能搜索系统知识溯源机制

Fig.6 Knowledge provenance mechanism in the Jiaodong gold deposit intelligent search system

图7 胶东金矿类型问题检索效果对比

Fig.7 Performance comparison on typical query types in Jiaodong gold deposit retrieval

表1 对比分析表(截止2025.03.10)

Table 1 Comparative analysis table (Data current through March 10, 2025)

山东六队提供准确答案	GraphRAG+大模型	ChatGPT-4o通用模型	DeepSeek	豆包
破碎带蚀变岩型	破碎带蚀变岩型	脉状金矿	破碎带蚀变岩型	破碎带蚀变岩型金矿
石英脉型	金石英脉型	层状金矿	层间滑动角砾岩型
硫化物石英脉型	硫化物石英脉型	接触带金矿	火山岩型
网脉型	破碎带石英网脉带型	浸染型金矿	石英脉型
盆缘断裂角砾岩型	盆缘断裂角砾岩型
蚀变砾岩型	蚀变砾岩型
层间滑脱拆离带型	层间滑动构造带型
黄铁矿碳酸盐脉型

表2 Ragas评估分析表

Table 2 RAGAS evaluation analysis

评判指标	RAG	GraphRAG
忠实度	0.94	0.96
语义相似度	0.87	0.89
上下文精度	0.86	0.96
上下文召回率	0.97	0.94

参考文献 40

[1]	周永章, 肖凡. 管窥人工智能与大数据地球科学研究新进展[J]. 地学前缘, 2024, 31(4): 1-6. DOI
[2]	王成彬, 王明果, 王博, 等. 融合知识图谱的矿产资源定量预测[J]. 地学前缘, 2024, 31(4): 26-36. DOI
[3]	ZHANG Y, WEI C, WU S, et al. Geogpt: understanding and processing geospatial tasks through an autonomous gpt[J]. arXiv preprint, 2023. DOI: arXiv:2307.07930.
[4]	马凯. 地质大数据表示与关联关键技术研究[D]. 武汉: 中国地质大学(武汉), 2018.
[5]	WANG B, MA K, WU L, et al. Visual analytics and information extraction of geological content for text-based mineral exploration reports[J]. Ore Geology Reviews, 2022, 144: 104818.
[6]	朱小龙. 地质文本中油气藏特征提取及成藏知识图谱构建研究[D]. 武汉: 中国地质大学(武汉), 2021.
[7]	杨立强, 邓军, 王中亮, 等. 胶东中生代金成矿系统[J]. 岩石学报, 2014, 30(9): 2447-2467.
[8]	冯雅杰, 王永志, 丁正江, 等. 胶西北金矿成矿模式及找矿技术方法演变[J]. 地学前缘, 2025, 32(4): 165-181.
[9]	王永志, 金樑, 朱月琴, 等. 基于大数据技术的地学文档关键词提取算法研发[J]. 地球物理学进展, 2018, 33(3): 1274-1281.
[10]	邱芹军, 田苗, 吴麒瑞, 等. 基于多源异构数据的地质知识图谱构建与应用[J/OL]. 地学前缘, 2024, 31: 1-17[2025-05-22]. https://doi.org/10.13745/j.esf.sf.2024.11.69.
[11]	CHEN J, LIN H, HAN X, et al. Benchmarking large language models in retrieval-augmented generation[C]// Proceedings of the AAAI Conference on Artificial Intelligence. 2024, 38(16): 17754-17762.
[12]	SIRIWARDHANA S, WEERASEKERA R, WEN E, et al. Improving the domain adaptation of retrieval augmented generation (RAG) models for open domain question answering[J]. Transactions of the Association for Computational Linguistics, 2023, 11: 1-17.
[13]	文森, 钱力, 胡懋地, 等. 基于大语言模型的问答技术研究进展综述[J]. 数据分析与知识发现, 2024, 8(6): 16-29. DOI
[14]	EDGE D, TRINH H, CHENG N, et al. From local to global: A graph rag approach to query-focused summarization[J]. arXiv preprint, 2024. DOI: arXiv:2404.16130.
[15]	MARKUS J. BUEHLERCA1. Generative retrieval-augmented ontologic graph and multiagent strategies for interpretive large language model-based materials design[J]. ACS Engineering Au, 2024, 4(2): 241-277. DOI PMID
[16]	PATRICK L, ETHAN P, ALEKSANDRA P, et al. Retrieval-augmented generation for knowledge-intensive NLP tasks[J]. Neural Information Processing Systems, 2020, 33: 9459-9474.
[17]	KHADIR A C, ALIANE H, GUESSOUM A. Ontology learning: grand tour and challenges[J]. Computer Science Review, 2021, 39: 100339.
[18]	周永章, 张前龙, 黄永健, 等. 钦杭成矿带斑岩铜矿知识图谱构建及应用展望[J]. 地学前缘, 2021, 28(3): 67-75. DOI
[19]	TORO S, ANAGNOSTOPOULOS A V, BELLO S M, et al. Dynamic retrieval augmented generation of ontologies using artificial intelligence (DRAGON-AI)[J]. Journal of Biomedical Semantics, 2024, 15(1): 19. DOI PMID
[20]	叶育鑫, 刘家文, 曾婉馨, 等. 基于本体指导的矿产预测知识图谱构建研究[J]. 地学前缘, 2024, 31(4): 16-25. DOI
[21]	HU X M, MA X G, MA C, et al. The geoscience knowledge system, ontology and knowledge graph for data-driven discovery: preface[J]. Geoscience Frontiers, 2023, 14(5): 332-333.
[22]	王岩, 王登红, 王成辉, 等. 基于地质大数据的中国金矿时空分布规律定量研究[J]. 地学前缘, 2024, 31(4): 438-455. DOI
[23]	刘向东, 丁正江, 邱昆峰, 等. 胶东台上—水旺庄巨型金矿床矿体空间分布规律及深部找矿预测[J/OL]. 地学前缘, 2024, 31: 1-19[2024-12-31]. https://doi.org/10.13745/j.esf.sf.2024.12.52.
[24]	王斌, 周明岭, 丁正江, 等. 胶西北蚀变岩型金矿构造控矿机制与三维深部定位预测[J]. 地学前缘, 2025, 32(4): 140-154.
[25]	张前龙, 周永章, 郭兰萱, 等. 找矿知识图谱的智能化应用: 以钦杭成矿带斑岩铜矿为例[J]. 地学前缘, 2024, 31(4): 7-15. DOI
[26]	DING Y J, FAN W Q, NING L B, et al. A survey on RAG meets LLMs: towards retrieval-augmented large language models[J]. arXiv preprint, 2024. DOI: arXiv:2405.06211v3.
[27]	WANG J, ZENG J, SHENG J. Enhancing and accelerating image-text retrieval with knowledge graphs and FAISS[C]// 2024 IEEE/WIC International Conference on Web Intelligence and Intelligent Agent Technology (WI-IAT). IEEE, 2024: 1-6.
[28]	GUPTA S, RANJAN R, SINGH S N. A comprehensive survey of Retrieval-Augmented Generation (RAG): evolution, current landscape and future directions[J]. arXiv preprint, 2024. DOI: arXiv:2410.12837.
[29]	FANG F T, BAI Y L, NI S W, et al. Enhancing noise robustness of retrieval-augmented language models with adaptive adversarial training[J]. arXiv preprint, 2024. DOI: arXiv:2405.20978.
[30]	DONG Y, WANG S, ZHENG H, et al. Advanced RAG models with graph structures: optimizing complex knowledge reasoning and text generation[C]// 2024 5th International Symposium on Computer Engineering and Intelligent Communications (ISCEIC). IEEE, 2024: 626-630.
[31]	HAN H Y, WANG Y, SHOMER H, et al. Retrieval-augmented generation with graphs (graphrag)[J]. arXiv preprint, 2024. DOI: arXiv:2501.00309.
[32]	韦一金, 任有强, 赵慧, 等. 基于GraphRAG的中国马铃薯新品种知识图谱构建[J]. 植物遗传资源学报, 2025, 26(6), 1229-1242.
[33]	WU J D, ZHU J Y, QI Y L, et al. Medical graph rag: towards safe medical large language model via graph retrieval-augmented generation[J]. arXiv preprint, 2024. DOI: arXiv:2408.04187.
[34]	PENG BC, ZHU Y, LIU Y C, et al. Graph retrieval-augmented generation: a survey[J]. arXiv preprint, 2024. DOI: arXiv:2408.08921v2.
[35]	ZHANG Q, CHEN S, BEI Y, et al. A Survey of graph retrieval-augmented generation for customized large language models[J]. arXiv preprint, 2025. DOI: arXiv:2501.13958.
[36]	QIU Q J, TIAN M, MA K, et al. A question answering system based on mineral exploration ontology generation: a deep learning methodology[J]. Ore Geology Reviews, 2023, 153: 105294.
[37]	宋明春, 李三忠, 伊丕厚, 等. 中国胶东焦家式金矿类型及其成矿理论[J]. 吉林大学学报(地球科学版), 2014, 44(1): 87-104.
[38]	宋明春. 胶东金矿深部找矿主要成果和关键理论技术进展[J]. 地质通报, 2015, 34(9): 1758-1771.
[39]	SARMAH B, MEHTA D, HALL B, et al. Hybridrag: integrating knowledge graphs and vector retrieval augmented generation for efficient information extraction[C]// Proceedings of the 5th ACM International Conference on AI in Finance. Brooklyn: New York University, 2024: 608-616.
[40]	ES S, JAMES J, ESPINOSA A L, et al. RAGAS: automated evaluation of retrieval augmented generation[C]// Proceedings of the 18th Conference of the European Chapter of the Association for Computational Linguistics:System Demonstrations. Malta: Association for Computational Linguistics, 2024: 150-158.

基于大模型与GraphRAG的胶东金矿智能搜索技术

Intelligent search technology for Jiaodong gold deposits based on large models and GraphRAG

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 9

参考文献 40

相关文章 15

编辑推荐

Metrics

本文评价

[1]	张志庭, 彭帅, 阙翔, 陈麒玉. 基于提示和度量学习的小样本地质关系抽取[J]. 地学前缘, 2025, 32(4): 250-261.
[2]	王斌, 周明岭, 丁正江, 张琪彬, 刘向东, 吕军阳, 周晓萍. 胶西北蚀变岩型金矿构造控矿机制与三维深部定位预测[J]. 地学前缘, 2025, 32(4): 140-154.
[3]	成秋明. 面向人类智能与人工智能融合的矿产资源预测新范式[J]. 地学前缘, 2025, 32(4): 1-19.
[4]	冯婷婷, 蔡诗柔, 张振杰. 基于知识图谱的碳酸岩型稀土矿成矿要素挖掘[J]. 地学前缘, 2025, 32(4): 262-279.
[5]	董宇浩, 王永志, 田江涛, 王成, 温世博, 李博文. 基于知识图谱的斑岩型铜矿预测研究进展[J]. 地学前缘, 2025, 32(4): 280-290.
[6]	王永志, 温世博, 李博文, 陈星宇, 董宇浩, 田江涛, 王斌, Muhammed Atif BILAL, 纪政, 孙丰月. 大模型驱动的矿产资源智能预测超级智能体构建方法探索[J]. 地学前缘, 2025, 32(4): 38-45.
[7]	朱彪彪, 曹伟, 虞鹏鹏, 张前龙, 郭兰萱, 原桂强, 韩枫, 王汉雨, 周永章. 基于CiteSpace的地质大数据与人工智能研究热点及前沿分析[J]. 地学前缘, 2024, 31(4): 73-86.
[8]	张前龙, 周永章, 郭兰萱, 原桂强, 虞鹏鹏, 王汉雨, 朱彪彪, 韩枫, 龙师尧. 找矿知识图谱的智能化应用:以钦杭成矿带斑岩铜矿为例[J]. 地学前缘, 2024, 31(4): 7-15.
[9]	季晓慧, 董雨航, 杨中基, 杨眉, 何明跃, 王玉柱. 基于知识图谱多跳推理的中文矿物知识问答方法与系统[J]. 地学前缘, 2024, 31(4): 37-46.
[10]	王成彬, 王明果, 王博, 陈建国, 马小刚, 蒋恕. 融合知识图谱的矿产资源定量预测[J]. 地学前缘, 2024, 31(4): 26-36.
[11]	叶育鑫, 刘家文, 曾婉馨, 叶水盛. 基于本体指导的矿产预测知识图谱构建研究[J]. 地学前缘, 2024, 31(4): 16-25.
[12]	周永章, 肖凡. 管窥人工智能与大数据地球科学研究新进展[J]. 地学前缘, 2024, 31(4): 1-6.
[13]	吕承训, 张达, 许亚青, 郭涛, 王宗永, 霍庆龙, 袁月蕾. 胶东金矿成矿深度的构造校正测算及成矿预测[J]. 地学前缘, 2022, 29(1): 427-438.
[14]	周永章, 张前龙, 黄永健, 杨威, 肖凡, 吉俊杰, 韩枫, 唐磊, 欧阳冲, 沈文杰. 钦杭成矿带斑岩铜矿知识图谱构建及应用展望[J]. 地学前缘, 2021, 28(3): 67-75.
[15]	毛先成，王迷军，刘占坤，陈进，邓浩. 基于勘查数据的胶东大尹格庄金矿床控矿地质因素定量分析[J]. 地学前缘, 2019, 26(4): 84-93.