An Explainable Rule-Based Framework for Knowledge Extraction and Legal Knowledge Graph Construction from Arabic Higher Education Legislation

Ghassan Khazal Ali  Khazal Ali

doi:10.69923/5hdbsy35

Authors

Ghassan University of Diyala Author https://orcid.org/0000-0003-4996-820X

DOI:

https://doi.org/10.69923/5hdbsy35

Keywords:

Knowledge graph, Knowledge Extraction, NLP

Abstract

The volume and complexity of legislative documents are increasing and have significant challenges for legal information management, retrieval, and intelligent decision support. Thus, the technologies related to natural language processing (NLP) and knowledge graph (KG) were used to improve the knowledge extraction in scientific and general domain text but are limited in Arabic legal documents. This limitation is evident in higher education legislation, where these documents have hierarchical structure, frequent changes, extensive cross-references, and special terminology. All of these factors make automated interpretation and knowledge extraction complicated. This work presents an automated framework for constructing an interpretable legal knowledge schema from the Iraqi Ministry of Higher Education legislation, using transparent information extraction based on Python. This framework was developed using a legislative dataset comprising 1930 regulations, laws, ministerial directives, and administrative decisions issued by the ministry of higher education between 1970 and 2025. The proposed work integrates legislative metadata with the full text of the article through many stages of pipelining, like document prepressing, segmentation into article and clause, legal entity identification, relegation extraction, and automatic graph construction. The resultant graph provides a structured representation of legislative knowledge that supports legal search, analysis, and exploration and serves as a foundational resource for future graph retrieval-augmented generation (GraphRAG) systems and legal question-answering applications. We conducted experiments on the complete legislative corpus, producing a legal knowledge graph comprising 10,312 article units, 16,597 clauses, 510 normalized entity nodes, and 12,289 extracted relation records.

Downloads

Download data is not yet available.

Author Biography

Ghassan, University of Diyala

Ghassan Khazal Ali is a lecturer at the College of Engineering, University of Diyala, Iraq. He received his B.Sc. degree in computer and software engineering from Al Mustansiriya University and his M.Sc. degree from UOITC, IRAQ. He holds a Ph.D. degree in information systems. His research areas are NLP and information retrieval, artificial intelligence, and software engineering. He has published several scientific papers in national and international conferences and journals. He can be contacted at email: ghassan_khazal@uodiyala.edu.iq.

References

[1] C. d’Amato, G. Rubini, F. Didio, D. Francioso, F. Z. Amara, and N. Fanizzi, “Automated creation of the legal knowledge graph addressing legislation on violence against women: Resource, methodology and lessons learned,” arXiv preprint arXiv:2508.06368, 2025. DOI: 10.48550/arXiv.2508.06368.

[2] S. Sirmokadam and S. Shahi, “A framework for analyzing legal documents by leveraging knowledge graphs,” in Data Management, Analytics and Innovation (2024), Lecture Notes in Networks and Systems, Springer Nature Singapore, 2024, pp. 417–425. DOI: 10.1007/978-981-97-3245-6_28.

[3] X. Zhu, H. Li, and T. Su, “Autonomous complex knowledge mining and graph representation through natural language processing and transfer learning,” Automation in Construction, vol. 155, Art. no. 105074, 2023. DOI: 10.1016/j.autcon.2023.105074.

[4] J. Li, L. Qian, P. Liu, and T. Liu, “Construction of legal knowledge graph based on knowledge-enhanced large language models,” Information, vol. 15, no. 11, Art. no. 666, 2024. DOI: 10.3390/info15110666.

[5] H. de Martim, “An ontology-driven Graph RAG for legal norms: A hierarchical, temporal, and deterministic approach,” in Legal Knowledge and Information Systems: JURIX 2025, 2025, pp. 282–287. DOI: 10.3233/FAIA251598.

[6] J. Chung, R. Ko, W. Yoo, M. Onizuka, S. Kim, T.-W. Kim, and W.-Y. Shin, “GraphCompliance: Aligning policy and context graphs for LLM-based regulatory compliance,” arXiv preprint arXiv:2510.26309, 2025. DOI: 10.48550/arXiv.2510.26309.

[7] I. Baimuratov and D. Turygin, “Representing normative regulations in OWL DL for automated compliance checking supported by text annotation,” arXiv preprint arXiv:2504.05951, 2025. DOI: 10.48550/arXiv.2504.05951.

[8] H. de Martim, “Deterministic legal agents: A canonical primitive API for auditable reasoning over temporal knowledge graphs,” arXiv preprint arXiv:2510.06002, 2025. DOI: 10.48550/arXiv.2510.06002.

[9] R. C. Barron et al., “Bridging legal knowledge and AI: Retrieval-augmented generation with vector stores, knowledge graphs, and hierarchical non-negative matrix factorization,” arXiv preprint arXiv:2502.20364, 2025. DOI: 10.48550/arXiv.2502.20364.

[10] E. Horner, C. Mateis, G. Governatori, and A. Ciabattoni, “Toward robust legal text formalization into defeasible deontic logic using LLMs,” arXiv preprint arXiv:2506.08899, 2025, doi: 10.48550/arXiv.2506.08899.

[11] E. Horner, C. Mateis, G. Governatori, and A. Ciabattoni, “Toward robust legal text formalization into defeasible deontic logic using LLMs,” arXiv preprint arXiv:2506.08899, 2025. DOI: 10.48550/arXiv.2506.08899.

[12] Iraqi Legislative Database, Supreme Judicial Council, Iraq, “Iraqi Legislative Database,” [Online]. Available: Iraqi Legislative Database. [Accessed: Aug. 31, 2026].

[13] "https://github.com/ghassankhazal79/LegalGRaph," [Online].

[14] X. Kang, L. Qu, L.-K. Soon, Z. Li, and A. Trakic, “Automating IRAC analysis in Malaysian contract law using a semi-structured knowledge base,” Artificial Intelligence and Law, 2025. DOI: 10.1007/s10506-025-09467-5.

[15] A. De Santis, M. Balduini, F. De Santis, A. Proia, A. Leo, M. Brambilla, and E. Della Valle, “Integrating large language models and knowledge graphs for extraction and validation of textual test data,” in The Semantic Web – ISWC 2024, Lecture Notes in Computer Science, vol. 15233, Springer, 2024, pp. 304–323. DOI: 10.1007/978-3-031-77847-6_17.