《计算机应用》唯一官方网站 ›› 2026, Vol. 46 ›› Issue (10): 3117-3123.DOI: 10.11772/j.issn.1001-9081.2025080945

• 人工智能 • 上一篇    

基于双通道嵌入融合的零样本知识图谱关系预测方法

高兵1, 林锦卓1, 邹启杰1(), 姚永强2   

  1. 1.大连大学 信息工程学院,辽宁 大连 116622
    2.大连大学附属中山医院,辽宁 大连 116001
  • 收稿日期:2025-08-19 修回日期:2025-10-07 接受日期:2025-10-16 发布日期:2025-11-18 出版日期:2026-10-10
  • 通讯作者: 邹启杰
  • 作者简介:高兵(1976—),男,黑龙江哈尔滨人,副教授,博士,CCF会员,主要研究方向:数据挖掘、大数据分析
    林锦卓(2001—),男,河北邢台人,硕士研究生,主要研究方向:知识图谱
    姚永强(1976—),男,辽宁大连人,教授,博士,主要研究方向:乳腺和甲状腺外科临床。
  • 基金资助:
    辽宁省科学研究经费资助项目(LJKZ1180);大连大学学科交叉一般类项目(DLUXK-2024-YB-002);大连大学学科建设专项(DLUXK-2025-FX-001)

Zero-shot knowledge graph relation prediction method via dual-channel embedding fusion

Bing GAO1, Jinzhuo LIN1, Qijie ZOU1(), Yongqiang YAO2   

  1. 1.College of Information Engineering,Dalian University,Dalian Liaoning 116622,China
    2.Affiliated Zhongshan Hospital of Dalian University,Dalian Liaoning 116001,China
  • Received:2025-08-19 Revised:2025-10-07 Accepted:2025-10-16 Online:2025-11-18 Published:2026-10-10
  • Contact: Qijie ZOU
  • About author:GAO Bing, born in 1976, Ph. D., associate professor. His research interests include data mining, big data analysis.
    LIN Jinzhuo, born in 2001, M. S. candidate. His research interests include knowledge graph.
    YAO Yongqiang, born in 1976, Ph. D., professor. His research interests include clinical breast and thyroid surgery.
  • Supported by:
    Scientific Research Project of Liaoning Province(LJKZ1180);Interdisciplinary General Project of Dalian University(DLUXK-2024-YB-002);Dalian University Discipline Development Special Project(DLUXK-2025-FX-001)

摘要:

零样本知识图谱关系预测任务旨在根据知识图谱已有的事实三元组推理预测包含未见关系的事实三元组。已有方法大多仅考虑关系的文本信息和图结构信息,而未充分利用实体的描述信息;针对关系的文本信息仅依赖于数据集本身的文本描述,难以捕捉关系的详细语义与知识图谱三元组整体的语义信息。为解决上述问题,本文提出一种基于双通道嵌入融合的零样本知识图谱关系预测方法。首先,使用结构编码器和语义编码器分别生成三元组实体部分的结构特征与语义特征;其次,通过双通道实体多模态信息融合模块融合得到的特征,同时利用大语言模型(LLM)的文本生成能力生成关系文本语义增强后的描述信息;再次,经过训练,语义增强后的语义特征与实体的多模态特征在向量空间中满足语义一致性;最后,针对零样本关系,根据语义增强后的文本描述信息完成零样本知识图谱关系预测。在两个真实数据集上的实验结果表明,相较于最优基线模型,在NELL_ZS数据集上,所提方法的平均倒数排名(MRR)、Hit@1、Hit@5和Hit@10分别提高了8.7、7.2、9.5和11.2个百分点;在WN18RR_ZSL数据集上,所提方法的MRR、Hit@1、Hit@5和Hit@10分别提高了6.6、9.2、3.9和0.8个百分点,验证了所提方法的有效性。

关键词: 知识图谱补全, 零样本知识图谱补全, 关系预测, 多模态融合, 生成对抗网络

Abstract:

Zero-shot knowledge graph relation prediction aims to infer triplets with unseen relations using the existing fact triplets. The existing methods often underutilize entity descriptions, only considering text information and graph structure information of relations. Text information for relations only relies on dataset-provided text descriptions, so it is difficult to capture the detailed semantics of relations and the overall semantic information of knowledge graph triplets. To address this issue, a zero-shot knowledge graph relation prediction method via dual-channel embedding fusion was proposed. First, a structural encoder and a semantic encoder were used to generate structural and semantic features of triplet entities. Then, the obtained features were fused via a dual-channel entity multimodal information fusion module, and the text generation ability of Large Language Models (LLMs) was used to generate description information after enhancing relation text semantics. Third, through training, semantically enhanced relation features and multimodal entity features were aligned semantically in vector space. Finally, for zero-shot relations, zero-shot knowledge graph relation prediction was realized according to semantically enhanced text description information. Experimental results on two real datasets show that compared to the optimal baseline method, on the NELL_ZS dataset, the proposed method improves Mean Reciprocal Rank (MRR), Hit@1, Hit@5, and Hit@10 by 8.7, 7.2, 9.5, and 11.2 percentage points, respectively; on the WN18RR_ZSL dataset, the proposed method improves MRR, Hit@1, Hit@5, and Hit@10 by 6.6, 9.2, 3.9, and 0.8 percentage points, respectively, demonstrating the effectiveness of the proposed method.

Key words: Knowledge Graph Completion (KGC), Zero-Shot Knowledge Graph Completion (ZS-KGC), relation prediction, multi-modal fusion, Generative Adversarial Network (GAN)

中图分类号: