Knowledge Graph Construction Based on Large Language Model and Its Application in Disaster Emergency
(School of Electronic Information, Huzhou College, Huzhou 313000, China)
Abstract
China is one of the countries with the most serious natural disasters in the world. Typhoons, earthquakes, floods and other disasters occur frequently, posing a huge threat to the safety of people's lives and property. Facing the complexity and urgency of disaster emergency management, Knowledge Graph technology has gradually become a research hotspot in this field by virtue of its advantages in information integration, relationship mining and knowledge reasoning. However, the heterogeneity and quality of disaster data are not uniform, which limits the ability of traditional Natural Language Processing methods to extract Knowledge Graph information. Therefore, this paper proposes a Large Language Model knowledge extraction method based on distributed prompt strategy. Through requirement hints, domain knowledge hints and few-shot hints, the Large Language Model can automatically extract entities and relationships from unstructured data, which improves the automation and accuracy of Knowledge Graph construction. In order to verify the effect of Knowledge Graph based on Large Language Model in practical application, this paper takes the typhoon disaster emergency plan as an example to construct the corresponding Knowledge Graph and its intelligent question and answer system.
Keywords
Knowledge Graph; Large Language Model; knowledge extraction; intelligent question and answer
References
- [1] He Haixia, Liu Tao, Du Ping Research on the Construction of Knowledge Graph for Earthquake Disaster Emergency Management [J]. Journal of Lanzhou Jiaotong University, 2023, 42 (3): 113-123
- [2] Li Zequan, Xu Shuhua, Li Bixiao, etc Disaster scenario information fusion technology based on knowledge graph [J]. Journal of North China University of Science and Technology, 2019, 16 (2): 1-5
- [3] YAN Y C,LIU L H,BAN Y K,et al. Dynamic Knowledge Graph Alignment [EB/OL].[2024-09-2].file:///C:/Users/wy/Downloads/16585-Article%20Text-20079-1-2-20210518.pdf.
- [4] ZHU Y Q,WANG X H,CHEN J,et al. Llms for Knowledge Graph Construction and Reasoning: Recent Capabilities and Future Opportunities [J].World Wide Web,2024,27(5):58-58.
- [5] ZHANG B W,SOH H. Extract, Define, Canonicalize: An LLm-Based Framework for Knowledge Graph Construction [J/OL].arXiv:2404.03868v1 [cs.CL].[2024-09-28].https://arxiv.org/html/2404.03868v1.
- [6] BROWN T B,MANN B,RYDER N,et al. Language Models are Few-Shot Learners [J/OL].arXiv:2005.14165 [cs.CL].[2024-09-28].https://arxiv.org/abs/2005.14165?context=cs.
- [7] ZHANG J Y,ZHANG Z X,ZHANG H H,et al. From Electronic Health Records to Terminology Base: A Novel Knowledge Base Enrichment Approach [J].Journal of Biomedical Informatics,2021,113:103628.
- [8] ZIAULLAH A W,OFLI F,IMRAN M. Monitoring Critical Infrastructure Facilities During Disasters Using Large Language Models [J/OL].arXiv:2404.14432 [cs.SI].[2024-09-28].https://arxiv.org/abs/2404.14432?context=cs.CL.
- [9] LIU Y C,KUO C L. Constructing Spatio-temporal Disaster Knowledge Graph from Social Media [J].AGILE:GIScience Series,2024,5:37.
- [10] Wang Zhe, Lu Junran, Yang Dongliang, etc Research on Intelligent Question and Answer System for Flood Emergency Decision Making Integrating GPT and Knowledge Graph [J]. China Safety Production Science and Technology, 2024, 20 (4): 5-11
- [11] OPENAI,ACHIAM J,ADLER S,et al. Gpt-4 Technical Report [J/OL].arXiv:2303.08774 [cs.CL].[2024-09-28].https://arxiv.org/abs/2303.08774?utm_source=chatgpt.com.
Funding
Huzhou University level Scientific Research Project (2024HXKM09)
About the authors
Xu Huan (2004-), male, Han ethnicity, from Hangzhou, Zhejiang, is currently pursuing a bachelor's degree. His research focuses on knowledge graph construction and enhancement techniques for large language models; Wu Mengfei (1995-), female, Han ethnicity, from Huzhou, Zhejiang, lecturer, Master of Engineering, research direction: natural language processing; Sun Wenxue (1989-), male, Han nationality, from Jining, Shandong, lecturer, Master of Engineering, research direction: intelligent computing, deep learning.