基于图结构与文本信息对比的知识图谱补全

doi:10.3778/j.issn.1002-8331.2407-0500

摘要/Abstract

摘要： 目前知识图谱补全方法主要使用图神经网络对图数据进行建模，忽视了文本信息在知识整合中的重要性。此外，现有模型未能充分挖掘细粒度三元组关系对实体嵌入更新的影响。针对以上问题，提出基于图结构与文本信息对比学习的知识图谱补全方法SCLKGC。在图编码器中采用特定类型的注意力机制，为实体邻域分配与其重要性相匹配的权重，从而实现实体邻域信息的更精准聚合。分别使用基于层次邻域和基于语义-结构的对比学习方法，以更有效地学习实体在知识图谱中的表示。实验部分使用两个真实数据集对所提方法进行验证。实验结果表明，与基线方法得到的最佳效果相比，在FB15k-237数据集上，SCLKGC的平均倒数排名（MRR）、Hits@1、Hits@3、Hits@10分别提高了0.008、0.016、0.011、0.013；在WN18RR数据集上，SCLKGC的MRR、Hits@1、Hits@3、Hits@10分别提高了0.006、0.008、0.006、0.012，验证了所提方法的有效性。

关键词: 知识图谱补全, 对比学习, 文本描述, 注意力机制

Abstract: Methods for knowledge graph completion primarily utilize graph neural networks to model graph data while neglecting the importance of textual semantics in knowledge integration. Existing models also fail to exploit the impact of fine-grained triple relationships on entity embeddings. This paper proposes a knowledge graph completion method based on structure and textual semantics contrastive learning, termed SCLKGC. A specific attention mechanism is employed in the graph encoder to allocate weights that match the importance of entities within their neighborhoods, achieving precise aggregation of neighborhood information. Hierarchical neighborhood-based and semantic-structural contrastive learning methods are utilized to effectively learn entity representations in the knowledge graph. Experiments validate the proposed method using two real-world datasets. Experimental results indicate that, compared to baseline methods, SCLKGC improves mean reciprocal rank (MRR), Hits@1, Hits@3, and Hits@10 by 0.008, 0.016, 0.011, 0.013, respectively, on the FB15k-237 dataset; on the WN18RR dataset, SCLKGC achieves improvements of 0.006, 0.008, 0.006, 0.012 in MRR, Hits@1, Hits@3, and Hits@10, respectively, validating the effectiveness of the proposed method.

Key words: knowledge graph completion, comparative learning, text description, attention mechanism

翟社平, 杨晴, 黄妍. 基于图结构与文本信息对比的知识图谱补全[J]. 计算机工程与应用, 2025, 61(17): 200-208.

ZHAI Sheping, YANG Qing, HUANG Yan. Knowledge Graph Completion Based on Comparison of Graph Structures and Text Semantics[J]. Computer Engineering and Applications, 2025, 61(17): 200-208.

参考文献

[1] 杜雪盈, 刘名威, 沈立炜, 等. 面向链接预测的知识图谱表示学习方法综述[J]. 软件学报, 2024, 35(1): 87-117.
DU X Y, LIU M W, SHEN L W, et al. Survey on represent-ation learning methods of knowledge graph for link predi-ction[J]. Journal of Software, 2024, 35(1): 87-117.
[2] 于梦波, 杜建强, 罗计根, 等. 基于知识表示学习的知识图谱补全研究进展[J]. 计算机工程与应用, 2023, 59(18): 59-73.
YU M B, DU J Q, LUO J G, et al. Research progress of knowledge graph completion based on knowledge representation learning[J]. Computer Engineering and Applications, 2023, 59(18): 59-73.
[3] BORDES A, USUNIER N, GARCIA-DURáN A, et al. Translating embeddings for modeling multi-relational data[C]//Proceedings of the 26th International Conference on Neural Information Processing Systems. New York: ACM, 2013: 2787-2795.
[4] NICKEL M, TRESP V, KRIEGEL H P. A three-way model for collective learning on multi-relational data [C]//Proceedings of the International Conference on Machine Learning. New York: JMLR, 2011: 3104482-3104584.
[5] VASHISHTH S, SANYAL S, NITIN V, et al. InteractE: impr-oving convolution-based knowledge graph embeddings by increasing feature interactions[C]//Proceedings of the AAAI Conference on Artificial Intelligence, 2020: 3009-3016.
[6] DAI G Q, WANG X Z, ZOU X Y, et al. MRGAT: multi-relational graph attention network for knowledge graph completion[J]. Neural Networks, 2022, 154: 234-245.
[7] YAO L, MAO C, LUO Y. KG-BERT: BERT for knowledge graph completion[J]. arXiv:1909.03193, 2019.
[8] WANG Z, ZHANG J W, FENG J L, et al. Knowledge graph embedding by translating on hyperplanes[C]//Proceedings of the AAAI Conference on Artificial Intelligence, 2014: 1112-1119.
[9] LIN Y, LIU Z, SUN M, et al. Learning entity and relation embeddings for knowledge graph completion[C]//Proceedings of the AAAI Conference on Artificial Intelligence, 2015: 2181-2187.
[10] XIAO H, HUANG M, HAO Y, et al. TransA: an adaptive approach for knowledge graph embedding[J]. arXiv:1509. 05490, 2015.
[11] SCHLICHTKRULL M, KIPF T N, BLOEM P, et al. Modeling relational data with graph convolutional networks[C]//Proceedings of the 15th International Conference on the Semantic Web. Cham: Springer, 2018: 593-607.
[12] WANG J, LI W, LIU F, et al. Hic-KGQA: improving multi-hop question answering over knowledge graph via hypergraph and inference chain[J]. Knowledge-Based Systems, 2023, 277: 110810.
[13] 张天成, 田雪, 孙相会, 等. 知识图谱嵌入技术研究综述[J]. 软件学报, 2023, 34(1): 277-311.
ZHANG T C, TIAN X, SUN X H, et al. Overview on knowledge graph embedding technology research[J]. Journal of Software, 2023, 34(1): 277-311.
[14] CAI H, ZHENG V W, CHANG K C C. A comprehensive survey of graph embedding: problems, techniques, and applications[J]. IEEE Transactions on Knowledge and Data Engineering, 2018, 30(9): 1616-1637.
[15] XIE R, LIU Z, JIA J, et al. Representation learning of knowledge graphs with entity descriptions[C]//Proceedings of the AAAI Conference on Artificial Intelligence, 2016: 29-35.
[16] WANG T, SHEN B, ZHONG Y. SSKGE: a time-saving knowledge graph embedding framework based on structure enhancement and semantic guidance[J]. Applied Intelligence, 2023, 53(21): 25171-25183.
[17] LIU X, WANG Z, SUN Y, et al. ISA-KGC: integrated semantics-structure analysis in knowledge graph completion[J]. IEEE Access, 2024, 12: 57250-57260.
[18] CHEN T, KORNBLITH S, NOROUZI M, et al. A simple framework for contrastive learning of visual representations[C]//Proceedings of the International Conference on Machine Learning, 2020: 1597-1607.
[19] WANG L, ZHAO W, WEI Z, et al. SimKGC: simple contrastive knowledge graph completion with pre-trained language models[C]//Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics, 2022: 4281-4294.
[20] WU S, SHEN X, XIA R. Commonsense knowledge graph completion via contrastive pre-training and node clustering[C]//Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics, 2023: 13977-13989.
[21] SHI G, ZHU Y, LIU J K, et al. HeGCL: advance self-supervised learning in heterogeneous graph-level representation[J]. IEEE Transactions on Neural Networks and Learning Systems, 2023: 1-12.
[22] TAN Z, CHEN Z, FENG S, et al. KRACL: contrastive learning with graph context modeling for sparse knowledge graph completion[C]//Proceedings of the ACM Web Conference, 2023: 2548-2559.
[23] 乔梓峰, 秦宏超, 胡晶晶, 等. 融合多视图对比学习的知识图谱补全算法[J]. 计算机科学与探索, 2024, 18(4): 1001-1009.
QIAO Z F, QIN H C, HU J J, et al. Knowledge graph completion algorithm with multi-view contrastive learning[J]. Journal of Frontiers of Computer Science and Technology, 2024, 18(4): 1001-1009.