BUILDING A KNOWLEDGE GRAPH ON NEO4J TO SUPPORT KNOWLEDGE RETRIEVAL FOR DATABASE COURSE
Keywords
DOI:
https://doi.org/10.34238/tnu-jst.15000Abstract
This paper proposes building a Knowledge Graph on the Neo4j platform to overcome the limitations of traditional retrieval methods (F1-score 57%) in performing multi-hop reasoning for the Database Fundamentals course. Entities and relationships were extracted from Vietnamese textbooks using rule-based pattern-matching rules, resulting in a knowledge graph composed of 70 concept nodes and 150 relational edges. The retrieval mechanism processes natural language into Cypher queries. An experimental evaluation on a set of 50 benchmark questions demonstrates that the system achieves an F1-score of 92% (an improvement of 35% compared to keyword-based lookup methods), an average response time of 150 milliseconds, and 88% accuracy for multi-hop inference. This research contributes a standardized Vietnamese Knowledge Graph dataset and a scalable multidimensional retrieval model for engineering education. These results indicate that integrating structured domain knowledge with graph-based retrieval and natural language-to-Cypher translation substantially enhances both the precision and reasoning capabilities of academic information access systems.Downloads
References
X. Zou, H. Lin, J. Wu, C. Zheng, and Q. Guan, “Constructing a knowledge graph for the database course group via deep learning,” in Proc. 2024 IEEE 8th Conf. Energy Internet Energy Syst. Integration (EI²), Changsha, China, Dec. 2024, pp. 1234-1240, doi: 10.1109/ICEIT61397.2024.10540737.
Y. Li, Y. Liang, R. Yang, J. Qiu, C. Zhang, and X. Zhang, “CourseKG: An educational knowledge graph based on course information for precision teaching,” Appl. Sci., vol. 14, no. 7, Apr. 2024, doi: 10.3390/app14072710.
B. Zhao et al., “EDUKG: A heterogeneous sustainable K-12 educational knowledge graph,” arXiv preprint arXiv:2210.12228, Oct. 2022. [Online]. Available: https://arxiv.org/abs/2210.12228. [Accessed Mar. 9, 2026].
K. Qu et al., “A survey of knowledge graph approaches and applications in education,” Electronics, vol. 13, no. 13, Jul. 2024, doi: 10.3390/electronics13132537.
A. Hogan et al., “Knowledge Graphs,” ACM Comput. Surv., vol. 54, no. 4, pp. 1-37, May 2021, doi: 10.1145/3447772.
F. Erxleben, M. Günther, M. Krötzsch, J. Mendez, and D. Vrandečić, “Introducing Wikidata to the Linked Data Web,” in Proc. 13th Int. Semantic Web Conf. (ISWC), Riva del Garda, Italy, 2014, pp. 50-65, doi: 10.1007/978-3-319-11964-9_4.
V. H. Nguyen et al., “Integrating knowledge based on ontology and knowledge graph for chatbot in education query,” HCMUE J. Sci., vol. 21, no. 5, pp. 771-785, May 2024, doi: 10.54607/hcmue.js.21.5.4129(2024).
K. T. Tran and M. H. Dinh, “Leveraging large-scale language models to efficiently transform natural language into Cypher queries,” (In Vietnamese), HUFLIT Journal of Science, vol. 9, no. 1, 2025, doi: 10.71091/2354-113X/248.
S. T. Sukumar, C.-H. Lung, M. Zaman, and R. Panday, “Knowledge Graph Generation and Application for Unstructured Data Using Data Processing Pipeline,” IEEE Access, vol. 12, pp. 136759-136770, Jan. 2024, doi: 10.1109/access.2024.3462635.
D. Diefenbach, V. López, K. Singh, and P. Maret, “Core Techniques of Question Answering Systems over Knowledge Bases: A Survey,” Knowl. Inf. Syst., vol. 55, no. 3, pp. 529-569, Jun. 2018, doi: 10.1007/s10115-017-1100-y.
Y. Zhang, Q. Chen, Z. Yang, H. Lin, and Z. Lu, “BioWordVec: Improving Biomedical Word Embeddings with Subword Information and MeSH,” Sci. Data, vol. 6, May 2019, doi: 10.1038/s41597-019-0055-0.
J. Devlin, M.-W. Chang, K. Lee, and K. Toutanova, “BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding,” in Proc. 2019 Conf. North Amer. Chapter Assoc. Comput. Linguist. (NAACL-HLT), Minneapolis, MN, USA, 2019, pp. 4171-4186, doi: 10.18653/v1/N19-1423.
