Page 240 - 《软件学报》2025年第12期

郑修林 等: 知识图谱补全技术及应用 5621

大成功. 但也因易产生幻觉和对捕获的事实知识缺乏可解释性而限制了其在医疗诊断等高风险场景中的应用. 传
统的 KGC 技术基于结构化知识表示和推理能够增强 LLM 获取知识的可解释性, 但是对新事实表示缺乏丰富的
语义, 而 LLM 能够提供丰富的语义表示. 因此, 大语言模型时代, KGC 技术研究仍然具有重要意义. 两者, 相辅相
成, 相互增强是当下及未来值得研究的课题.
9 结 论

本文根据模型构建过程所需样本的数量对已有的 KGC 模型进行了分类, 即零样本 KGC 模型、少样本 KGC
模型和多样本 KGC 模型, 并对每类模型进行了全面的分析和讨论, 指出了它们的优点和不足, 并通过实验加以验
证. 同时例举 KGC 技术的一般应用和华谱系统中的具体实现. 基于以上调研, 本文总结了当下 KGC 模型的不足
和展望了 KGC 未来可能的研究方向, 以期对未来 KGC 研究提供一些基本的借鉴和参考.

References:
[1] Steiner T, Verborgh R, Troncy R, Gabarró J, van de Walle R. Adding realtime coverage to the Google knowledge graph. In: Proc. of the
2012 Int’l Conf. on Posters & Demonstrations Track. Boston: CEUR-WS.org, 2012. 65–68.
[2] Liu K, Zhao J, He SZ, Zhang YZ. Question answering over knowledge bases. IEEE Intelligent Systems, 2015, 30(5): 26–35. [doi: 10.
1109/MIS.2015.70]
[3] Guo QM, Wang X, Zhu ZF, Liu PY, Xu LC. A knowledge inference model for question answering on an incomplete knowledge graph.
Applied Intelligence, 2023, 53(7): 7634–7646. [doi: 10.1007/s10489-022-03927-0]
[4] Bi X, Nie HJ, Zhao XG, Yuan Y, Wang GR. Reinforcement learning inference techniques for knowledge graph constrained question
answering. Ruan Jian Xue Bao/Journal of Software, 2023, 34(10): 4565–4583 (in Chinese with English abstract). http://www.jos.org.cn/
1000-9825/6889.htm [doi: 10.13328/j.cnki.jos.006889]
[5] Qiao SJ, Yang GP, Yu Y, Han N, Tan X, Qu LL, Ran LQ, Li H. QA-KGNet: Language model-driven knowledge graph question-
answering model. Ruan Jian Xue Bao/Journal of Software, 2023, 34(10): 4584–4600 (in Chinese with English abstract). http://www.jos.
org.cn/1000-9825/6882.htm [doi: 10.13328/j.cnki.jos.006882]
[6] Jannach D, Manzoor A, Cai WL, Chen L. A survey on conversational recommender systems. ACM Computing Surveys (CSUR), 2022,
54(5): 105. [doi: 10.1145/3453154]
[7] Wang HW, Zhao M, Xie X, Li WJ, Guo MY. Knowledge graph convolutional networks for recommender systems. In: Proc. of the 2019
World Wide Web Conf. San Francisco: ACM, 2019. 3307–3313. [doi: 10.1145/3308558.3313417]
[8] Li LS, Dong JY, Qin XY. Dual-view graph neural network with gating mechanism for entity alignment. Applied Intelligence, 2023,
53(15): 18189–18204. [doi: 10.1007/s10489-022-04393-4]
[9] Ji SX, Pan SR, Cambria E, Marttinen P, Yu PS. A survey on knowledge graphs: Representation, acquisition, and applications. IEEE
Trans. on Neural Networks and Learning Systems, 2022, 33(2): 494–514. [doi: 10.1109/TNNLS.2021.3070843]
[10] Zhang X, Zhang CX, Guo JT, Peng C, Niu ZD, Wu XD. Graph attention network with dynamic representation of relations for
knowledge graph completion. Expert Systems with Applications, 2023, 219: 119616. [doi: 10.1016/j.eswa.2023.119616]
[11] Li WD, Peng R, Li Z. Knowledge graph completion by jointly learning structural features and soft logical rules. IEEE Trans. on
Knowledge and Data Engineering, 2023, 35(3): 2724–2735. [doi: 10.1109/TKDE.2021.3108224]
[12] Zhang NY, Xie X, Chen X, Deng SM, Ye HB, Chen HJ. Knowledge collaborative fine-tuning for low-resource knowledge graph
completion. Ruan Jian Xue Bao/Journal of Software, 2022, 33(10): 3531–3545 (in Chinese with English abstract). http://www.jos.org.cn/
1000-9825/6628.htm [doi: 10.13328/j.cnki.jos.006628]
[13] Wang Q, Wang B, Guo L. Knowledge base completion using embeddings and rules. In: Proc. of the 24th Int’l Conf. on Artificial
Intelligence. Buenos Aires: AAAI Press, 2015. 1859–1865.
[14] Wang Q, Mao ZD, Wang B, Guo L. Knowledge graph embedding: A survey of approaches and applications. IEEE Trans. on Knowledge
and Data Engineering, 2017, 29(12): 2724–2743. [doi: 10.1109/TKDE.2017.2754499]
[15] Rossi A, Barbosa D, Firmani D, Matinata A, Merialdo P. Knowledge graph embedding for link prediction: A comparative analysis.
ACM Trans. on Knowledge Discovery from Data (TKDD), 2021, 15(2): 14. [doi: 10.1145/3424672]
[16] Shen T, Zhang F, Cheng JW. A comprehensive overview of knowledge graph completion. Knowledge-based Systems, 2022, 255:
109597. [doi: 10.1016/j.knosys.2022.109597]