A Study on Large-Scale Model Knowledge Extraction Methods for Enterprise-Level Data Governance

Yankun Li; Meng Wu; Longfei Ren; Zhiheng Guo; Wei Zhang

doi:10.63313/EPP.9009

Authors

Yankun Li China United Network Communications Corporation Limited Software Research Institute, Beijing, 100176, China Author
Meng Wu China United Network Communications Corporation Limited Software Research Institute, Beijing, 100176, China Author
Longfei Ren China United Network Communications Corporation Limited Software Research Institute, Beijing, 100176, China Author
Zhiheng Guo China United Network Communications Corporation Limited Software Research Institute, Beijing, 100176, China Author
Wei Zhang China United Network Communications Corporation Limited Software Research Institute, Beijing, 100176, China Author

DOI:

https://doi.org/10.63313/EPP.9009

Keywords:

Data governance, Knowledge extraction, Knowledge graph, Deep learning, BERT-BiLSTM-CRF

Abstract

Enterprise data governance demands precise and standardized approaches to knowledge management. Addressing the limitations of traditional knowledge extraction methods when processing enterprise data governance documents, this study proposes an improved BERT-BiLSTM-CRF-based knowledge extraction method. It designs and implements a multi-level knowledge base architecture with an incremental update mechanism. A knowledge application service framework tailored for data governance is constructed, enabling intelligent decision support in core scenarios such as data standard formulation and quality diagnosis. This provides an effective technical solution for enterprise data governance.

References

[1] Ling W . Application of Large Language Models in Data Governance in the Petroleum and Petrochemical Industry[J].China Oil & Gas, 2024, 31(2):27-32.

[2] Yiran C , Zhengyin H U , Chunjiang L .Analysis of Progress in Data Mining of Scientific Liter-ature Using Large Language Models[J].Journal of Library & Information Science in Agri-culture, 2025, 37(2).

[3] Liu Y , Ma S , Yang Z ,et al.Domain knowledge-assisted materials data anomaly detection towards constructing high-performance machine learning models[J].Journal of Materi-omics, 2025, 11(6).

[4] Minghao W , Jiexin H , Jiabin C ,et al.Research on the technical route of standards digitiza-tion based on knowledge graphs[J].China Standardization, 2024(S1):26-33.

[5] Naskar A , Goel T , Chauhan V ,et al.Knowledge-driven neural models for extraction and analysis of sustainability events from the web[J].Environmental Data Science, 2024, 3(000).

[6] Serderidis K , Konstantinidis I , Meditskos G ,et al.d2kg: An integrated ontology for knowledge graph-based representation of government decisions and acts[J].SEMANTIC WEB, 2024, 15(5).

[7] Zhang C , Ji Y , Yan Y .An Introduction to the Implementation Strategy of Unstructured Data Governance for Aviation Enterprise[J].2023 9th International Conference on Big Data and Information Analytics (BigDIA), 2023:350-355.

[8] Angioni S , Consoli S ,Danilo Dessì,et al.Exploring Environmental, Social, and Governance (ESG) Discourse in News: An AI-Powered Investigation Through Knowledge Graph Analy-sis[J].IEEE Access, 2024, 12(000):15.

[9] Tibble H , Alyami R A , Bush A ,et al.Using routine primary care data in research: (in)efficient case studies and perspectives from the Asthma UK Centre for Applied Re-search[J].BMJ HEALTH & CARE INFORMATICS, 2025, 32(1).

[10] Basereh M , Caputo A , Brennan R .Automatic transparency evaluation for open knowledge extraction systems[J].Journal of Biomedical Semantics, 2023, 14(1).