A Knowledge-Graph-Based Intelligent Agent for Domain-Specific Question Answering

Mengxue Zhang, Wangwu Huang

Abstract

This research focuses on the development of a question-answering intelligent agent for specialized domains based on a knowledge graph, using the field of English linguistics as an example. English linguistics encompasses a vast array of conceptual terminology and a complex knowledge structure that often lacks sufficient concrete language examples for illustration. Traditional information retrieval methods fall short in meeting the advanced demands of semantic comprehension and knowledge inference required in linguistics education. To address these challenges, this study employs Graph Retrieval-Augmented Generation (GraphRAG) technology to design and implement an intelligent agent on the Wenxin Agent Platform. This agent integrates large language models with structured knowledge graphs to enhance the performance of real-time question-answering. The knowledge graph functions as a structured repository that organizes domain-specific linguistic knowledge, thereby expanding the agent's knowledge reserve and enabling more precise responses that facilitate students' systematic understanding of linguistic concepts. Evaluation results indicate that the proposed agent surpasses general-purpose models in response accuracy and professionalism, depth and logical coherence, and explanatory transparency, while effectively addressing hallucination issues common in conventional large language models. The system demonstrates enhanced domain adaptability and improved reasoning performance within linguistic contexts.

Keywords

linguistics; GraphRAG; intelligent agents; knowledge graph; Large Language Models

Full Text:

PDF

References

Allemang, D., & Sequeda, J. (2024). Increasing the LLM accuracy for question answering: Ontologies to the rescue! arXiv. https://doi.org/10.48550/arXiv.2405.11706

Brown, T. B., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., ... & Amodei, D. (2020). Language models are few-shot learners. arXiv. https://doi.org/10.48550/arXiv.2005.14165

Buehler, M. J. (2024). Generative retrieval-augmented ontologic graph and multiagent strategies for interpretive large language model-based materialsdesign. ACS Engineering Au, 4(2), 241–277. https://doi.org/10.1021/acsengineeringau.3c00063

Chen, J., Lin, H., Han, X., Liang, H., Wu, W., & Shi, H. (2024). Benchmarking large language models in retrieval-augmented generation. Proceedings of the AAAI Conference on Artificial Intelligence, 38(16), 17754–17762. AAAI Press.

Dash, D., Thapa, R., Banda, J. M., Swaminathan, A., Cheatham, M., & Kaggal,V. (2023). Evaluation of GPT-3.5 and GPT-4 for supporting real-world information needs in healthcare delivery. arXiv. https://doi.org/10.48550/arXiv.2304.13714

Dong, Z. A., Qin, K. H., Zhou, Z. L., Wang, S. F., Xiong, J., & Liu, Z. B. (2025). GraphRAG-based question answering system for traditional Chinese medicine knowledge. Journal of Qufu Normal University (Natural Science Edition), 1–11.

Fan, X. S., Li, Z. Q., Zhang, X. H., & Wang, S. L. (2025). A Q&A system forsteel industry standards based on large models and knowledge graph-enhanced RAG. Metallurgical Automation, 49(4), 134–145.

Gao, L. P., Cai, W. J., Zheng, W., Zheng, J. B., & Ma, C. Y. (2025). Application and practice of an intelligent tutoring system based on large models and knowledge graphs in blended teaching of software testing courses. Software Guide, 1–8.

Jiang, X., Wang, H. Y., Yang, S. T., Chen, X. Y., Zhou, M. Y., & Dong, L. G. (2025). Research on multi-hop question answering technology based on the integration of large language models and knowledge graphs. Telecommunications Science, 1–25.

Lewis, P., Perez, E., Piktus, A., Petroni, F., Karpukhin, V., Goyal, N., ... & Kiela, D. (2020). Retrieval-augmented generation for knowledge-intensive NLP tasks. Advances in Neural Information Processing Systems, 33, 9459–9474.

Li, B. W., Wang, Y. Z., Ding, Z. J., Wang, B., Wen, S. B., Dong, Y. H., & Ji, Z.(2025). Intelligent search technology for Jiaodong gold deposits based on large models and GraphRAG. Earth Science Frontiers, 32(4), 155–164. https://doi.org/10.13745/j.esf.sf.2025.4.77

Ma, K. (2018). Research on key technologies of geological big data representation and association [Doctoral dissertation, China University of Geosciences]. CNKI. https://kns.cnki.net/kcms2/article/abstract?v=9oehDy4zW5YO2ZN3AfUGiOSiMDg3yk7NulMyYZDKHndWnqU7ftlZzWDmDNf2Mt166oEd9pIHnZlMNcI2bxif6oVNFK_-qhKKgOH9sA3MLJzLcdg67t1EfNtw6dQY05clX08s0A2CB0RhH3tZWkSQGtVHIPLadrakvyjWNkc9lBDq7U-ZuYMYGNQMgiHS-_kG

Ma, W. L., Sun, T., Zhao, R. X., & Xian, G. J. (2025). Generating graphical summaries of scientific literature based on LLM and GraphRAG. Data Analysis and Knowledge Discovery, 1–21.

Ministry of Education of the People's Republic of China. (2022). Compulsory education English curriculum standards (2022 edition). Beijing Normal University Publishing Group.

Peng, B., Zhu, Y., Liu, Y., Li, H., & Zeng, M. (2024). Graph retrieval-augmented generation: A survey. arXiv. https://doi.org/10.48550/arXiv.2408.08921

Siddharth, L., & Luo, J. (2024). Retrieval augmented generation using engineering design knowledge. Knowledge-Based Systems, 303, 112410. https://doi.org/10.1016/j.knosys.2024.112410

Siriwardhana, S., Weerasekera, R., Wen, E., & Kaluarachchi, T. (2023). Improving the domain adaptation of retrieval augmented generation (RAG) models for open domain question answering. Transactions of the Association for Computational Linguistics, 11, 1–17. https://doi.org/10.1162/tacl_a_00524

Wang, B., Ma, K., Wu, L., & Wang, C. (2022). Visual analytics and information extraction of geological content for text-based mineral exploration reports. Ore Geology Reviews, 144, 104818. https://doi.org/10.1016/j.oregeorev.2022.104818

Wang, C. B., Wang, M. G., Wang, B., Chen, J. G., Ma, X. G., & Jiang, S. (2024). Quantitative prediction of mineral resources integrating knowledge graphs. Earth Science Frontiers, 31(4), 26–36. https://doi.org/10.13745/j.esf.sf.2024.5.3

Wang, H., Yan, W., Zhang, X. M., Zhu, S., Jiang, Z. G., & Zhu, Z. R. (2025). Disassembly sequence planning for end-of-life power batteries based on GraphRAG. China Mechanical Engineering, 1–15.

Wang, R. (2024). The new paradigm of AI-empowered continuing education in universities. Continuing Education Research, 3, 12–16.

Wei, Y. D. (2024). The "cognitive turn" of artificial intelligence: The necessary path to artificial general intelligence. Journal of Shanxi Normal University (Social Science Edition), 51(5), 9–17. https://doi.org/10.16207/j.cnki.1001-5957.2024.05.010

Wen, S., Qian, L., Hu, M. D., & Chang, Z. J. (2024). A review of question answering technology based on large language models. Data Analysis and Knowledge Discovery, 8(6), 16–29.

Xu, S., Tong, J. R., & Hu, X. E. (2024). Next-generation personalized learning: Generative artificial intelligence-enhanced intelligent tutoring systems. Open Education Research, 30(2), 13–22. https://doi.org/10.13966/j.cnki.kfjyyj.2024.02.002

Zheng, W., Gu, R., Wu, X., & Ma, C. (2023). Personalized learning path recommendations for software testing courses based on knowledge graphs. Computer Education, 12, 63–70.

Zhong, M. (2025). Building an AI agent spoken language practice model based on characters from English textbooks. Teaching and Administration, 11,55–58.

Zhou, Y. Z., & Xiao, F. (2024). A glimpse into new advances in artificial intelligence and big data earth science research. Earth Science Frontiers, 31(4), 1–6. https://doi.org/10.13745/j.esf.sf.2024.6.99

DOI: https://doi.org/10.20961/transling.v5i3.109928

Refbacks

  • There are currently no refbacks.