基于大模型验证增强的产业链知识图谱构建研究

doi:10.3778/j.issn.1673-9418.2505073

摘要/Abstract

摘要： 随着产业链分析的深入和信息化需求的增长，如何从海量文本中高效抽取企业及产业间的供需关系并构建产业链知识图谱，成为当前研究的主要问题。基于此，提出了一种基于大语言模型验证增强的产业链知识图谱构建方法。在概念层，设计了以“产业-企业”双视角的图谱构建层级体系，实现企业与产业的映射关系。在数据层，对结构化数据利用大语言模型进行知识抽取，构建产业链标注语料库（CERIC）。对于非结构化数据，设计了一个验证增强的大模型数据抽取框架VRTE-LLM。该框架使用CERIC语料库对大语言模型微调训练，以提升其在特定领域的知识理解与信息抽取能力，并根据任务需求和文本上下文，修正关系三元组中的完整性错误与偏差错误，通过使用预定义的规则库对三元组进行验证。实验结果显示，该框架在实体识别与关系抽取任务中分别达到80.9%与83.9%的准确率。所构建的产业链知识图谱包含39 627个三元组，涵盖四大产业领域、70余家企业的知识图谱，覆盖产业链上下游各关键环节，为产业链进行量化分析、趋势预测提供技术支持。

关键词: 知识图谱构建, 命名实体识别, 关系三元组抽取, 大语言模型（LLM）, 产业链

Abstract: With the deepening of industrial chain analysis and the increasing demand for information processing, efficiently extracting supply-demand relationships between enterprises and industries from massive text data and constructing industrial chain knowledge graphs have become a critical research issue. To address this challenge, this paper proposes a knowledge graph construction method for industrial chains enhanced by large language model verification. At the conceptual level, this paper designs a hierarchical architecture for graph construction from dual perspectives of “industry-enterprise” to establish mapping relationships between enterprises and industrial sectors. At the data level, structured data undergoes knowledge extraction through large language models to develop an industrial chain domain corpus for entity and relation annotation (corpus for entity and relation annotation in industrial chain domain, CERIC). For unstructured data, this paper implements a verification-augmented relation triple extraction LLM (VRTE-LLM), which employs the CERIC corpus to fine-tune the large language model, thereby enhancing domain-specific knowledge comprehension and information extraction capabilities. The framework systematically rectifies completeness errors and deviation errors in relational triples by integrating contextual analysis with task-specific requirements, while applying a pre-defined rule base for triple validation to significantly improve the accuracy of enterprise information identification from unstructured texts. Experimental results demonstrate that the framework achieves 80.9% accuracy in entity recognition and 83.9% accuracy in relation extraction. The constructed industrial chain knowledge graph comprises over 39627 triples, covering four major industrial domains and knowledge representations of more than 70 enterprises, with comprehensive mapping across critical upstream and downstream chain segments. This methodology provides technical support for quantitative analysis and trend prediction in industrial chain research.

Key words: knowledge graph construction, named entity recognition, relational triple extraction, large language model (LLM), industry chain

郑傲泽, 张坤丽, 李云龙, 王影, 袁颂瑞, 吴鹏程, 贾玉祥, 昝红英. 基于大模型验证增强的产业链知识图谱构建研究[J]. 计算机科学与探索, 2025, 19(12): 3328-3339.

ZHENG Aoze, ZHANG Kunli, LI Yunlong, WANG Ying, YUAN Songrui, WU Pengcheng, JIA Yuxiang, ZAN Hongying. Construction of Industry Chain Knowledge Graph via Large Language Model-Based Verification Enhancement[J]. Journal of Frontiers of Computer Science and Technology, 2025, 19(12): 3328-3339.

参考文献

[1] 龚勤林. 产业链延伸的价格提升研究[J]. 价格理论与实践, 2004(3): 33-34.
GONG Q L. Research on price increase through industrial chain extension[J]. Price: Theory & Practice, 2004(3): 33-34.
[2] 严珂, 王红姝. 基于产业链视角的速冻食品产业发展研究: 以河南速冻食品行业为例[J]. 经济师, 2015(2): 57-59.
YAN K, WANG H S. Research on the development of the frozen food industry from the industrial chain perspective: a case study of the frozen food industry in Henan[J]. China Economist, 2015(2): 57-59.
[3] 周月书, 王婕. 产业链组织形式、市场势力与农业产业链融资: 基于江苏省397户规模农户的实证分析[J]. 中国农村经济, 2017(4): 46-58.
ZHOU Y S, WANG J. Industry chain organization mode, market power and agricultural industry chain financing: an empirical analysis based on data collected from 397 large-scale farmers in Jiangsu Province[J]. Chinese Rural Economy, 2017(4): 46-58.
[4] CHEN X J, JIA S B, XIANG Y. A review: knowledge reasoning over knowledge graph[J]. Expert Systems with Applications, 2020, 141: 112948.
[5] HUBAUER T, LAMPARTER S, HAASE P, et al. Use cases of the industrial knowledge graph at Siemens[C]//Proceedings of the 2018 International Semantic Web Conference, 2018: 107-108.
[6] YAO Y F, DUAN J H, XU K D, et al. A survey on large langu-age model (LLM) security and privacy: the good, the bad, and the ugly[J]. High-Confidence Computing, 2024, 4(2): 100211.
[7] HE W, NIU Y, DAI S, et al. Study on the system method of DCS remote operation and maintenance system of nuclear power plant based on parallel theory[EB/OL]. [2025-03-10]. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4679098.
[8] 李逸飞, 张玲玲, 董宇轩, 等. 基于大语言模型增强表征对齐的小样本持续关系抽取方法[J]. 计算机科学与探索, 2024, 18(9): 2326-2336.
LI Y F, ZHANG L L, DONG Y X, et al. Large language model augmentation and feature alignment method for few-shot continual relation extraction[J]. Journal of Frontiers of Computer Science and Technology, 2024, 18(9): 2326-2336.
[9] HOGAN A, BLOMQVIST E, COCHEZ M, et al. Knowledge graphs[J]. ACM Computing Surveys, 2022, 54(4): 1-37.
[10] NGUYEN H L, VU D T, JUNG J J. Knowledge graph fusion for smart systems: a survey[J]. Information Fusion, 2020, 61: 56-70.
[11] LIN J J, ZHAO Y Z, HUANG W Y, et al. Domain knowledge graph-based research progress of knowledge representation[J]. Neural Computing and Applications, 2021, 33(2): 681-690.
[12] EHRMANN M, HAMDI A, PONTES E L, et al. Named entity recognition and classification in historical documents: a survey[J]. ACM Computing Surveys, 2024, 56(2): 1-47.
[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.
WEN S, QIAN L, HU M D, et al. Review of research progress on question-answering techniques based on large language models[J]. Data Analysis and Knowledge Discovery, 2024, 8(6): 16-29.
[14] WU T Y, HE S Z, LIU J P, et al. A brief overview of ChatGPT: the history, status quo and potential future development[J]. IEEE/CAA Journal of Automatica Sinica, 2023, 10(5): 1122-1136.
[15] SHEN Y, SONG K, TAN X, et al. HuggingGPT: solving AI tasks with ChatGPT and its friends in hugging face[EB/OL]. [2025-03-10]. https://arxiv.org/abs/2303.17580.
[16] 裴炳森, 李欣, 吴越. 基于ChatGPT的电信诈骗案件类型影响力评估[J]. 计算机科学与探索, 2023, 17(10): 2413-2425.
PEI B S, LI X, WU Y. Influence evaluation of telecom fraud case types based on ChatGPT[J]. Journal of Frontiers of Computer Science and Technology, 2023, 17(10): 2413-2425.
[17] 冯钧, 畅阳红, 陆佳民, 等. 基于大语言模型的水工程调度知识图谱的构建与应用[J]. 计算机科学与探索, 2024, 18(6): 1637-1647.
FENG J, CHANG Y H, LU J M, et al. Construction and application of knowledge graph for water engineering scheduling based on large language model[J]. Journal of Frontiers of Computer Science and Technology, 2024, 18(6): 1637-1647.
[18] WANG S H, SUN X F, LI X Y, et al. GPT-NER: named entity recognition via large language models[EB/OL]. [2025-03-10]. https://arxiv.org/abs/2304.10428.
[19] WEBSON A, PAVLICK E. Do prompt-based models really understand the meaning of their prompts?[EB/OL]. [2025-03-10]. https://arxiv.org/abs/2109.01247.
[20] JIANG X J, ZHOU W, HOU J. Construction of fault diagnosis system for control rod drive mechanism based on knowledge graph and Bayesian inference[J]. Nuclear Science and Techniques, 2023, 34(2): 21.
[21] FAN C H, WEI W, QU X Y, et al. Enhancing low-resource relation representations through multi-view decoupling[J]. Proceedings of the AAAI Conference on Artificial Intelligence, 2024, 38(16): 17968-17976.
[22] YUAN J S, LI H Q. Research on the standardization model of data semantics in the knowledge graph construction of Oil&Gas industry[J]. Computer Standards & Interfaces, 2023, 84: 103705.
[23] 魏明珠, 郑荣, 高志豪, 等. 融合知识图谱和深度神经网络的产业新兴技术预测模型研究[J]. 情报学报, 2022, 41(11): 1134-1148.
WEI M Z, ZHENG R, GAO Z H, et al. Research on industry emerging technology forecast modeling based on knowledge graph and deep neural networks[J]. Journal of the China Society for Scientific and Technical Information, 2022, 41(11): 1134-1148.
[24] 刘家玮, 刘波, 沈岳. 知识图谱在农业信息服务中的应用进展[J]. 软件, 2015, 36(3): 26-30.
LIU J W, LIU B, SHEN Y. The application of the knowledge mapping based on agricultural information services[J]. Computer Engineering & Software, 2015, 36(3): 26-30.
[25] 范存庆, 余军合, 战洪飞, 等. 产业集群知识图谱构建方法研究[J]. 科技与经济, 2022, 35(3): 56-60.
FAN C Q, YU J H, ZHAN H F, et al. Research on the construction method of industrial cluster knowledge graph[J]. Science & Technology and Economy, 2022, 35(3): 56-60.
[26] ZHANG W Y, LIU Y G, JIANG L H, et al. The construction of a domain knowledge graph and its application in supply chain risk analysis[C]//Advances in E-Business Engineering for Ubiquitous Computing. Cham: Springer, 2020: 464-478.
[27] WEI X, CUI X Y, CHENG N, et al. ChatIE: zero-shot information extraction via chatting with ChatGPT[EB/OL]. [2025-03-11]. https://arxiv.org/abs/2302.10205.
[28] YANG A, YANG B, ZHANG B, et al. Qwen2.5 technical report[EB/OL]. [2025-03-11]. https://arxiv.org/abs/2412.15115.
[29] GUO D, YANG D, ZHANG H, et al. DeepSeek-R1: incentivizing reasoning capability in LLMs via reinforcement learning[EB/OL]. [2025-03-11]. https://arxiv.org/abs/2501.12948.
[30] WEI Z P, SU J L, WANG Y, et al. A novel cascade binary tagging framework for relational triple extraction[EB/OL]. [2025-03-11]. https://arxiv.org/abs/1909.03227.
[31] LU Y, LIU Q, DAI D, et al. Unified structure generation for universal information extraction[EB/OL]. [2025-03-11]. https://arxiv.org/abs/2203.12277.