结合实体和关系消息传递的低资源知识图谱补全

doi:10.3778/j.issn.1002-8331.2312-0151

摘要/Abstract

摘要： 知识图谱补全是知识图谱领域的一个重要研究问题。现有的知识图谱补全工作大多假设在充足的三元组实例上进行，然而在医疗、法律等垂直领域，数据难以获取，缺乏足够的数据和先验知识，属于低资源场景。研究低资源场景下的知识图谱补全方法对于解决实际问题具有重要意义。提出结合节点和边消息传递的低资源知识图谱补全方法SMKGC（similarity messages knowledge graph completion），该方法通过感知局部关系（边）与实体（节点）实现对链接的预测。与现有方法不同，SMKGC将语义相似节点及语义相似边的特征信息与Pathcon模型进行融合，从而通过增强消息传递的特征表示来提升链接预测的准确性。具体包括两个模块：（1）基于实体相似的消息传递，在捕获给定实体对相邻实体的基础上，聚合周围相似实体信息；（2）基于关系相似的消息传递，通过关系路径获取给定实体对的相对位置，同时结合路径的相似边实现关系预测。实验结果表明，该方法在知识图谱常用的基准数据集上明显优于其他方法，同时也证明了结合相似实体和关系消息传递在低资源知识图谱补全任务中可以有效提升预测的准确度。

关键词: 知识图谱, 相似性, 链接预测

Abstract: Knowledge graph completion is a significant research topic in knowledge graphs. Most existing work assumes that it is carried out on sufficient triadic instances. However, in vertical areas such as medicine and law, data are difficult to obtain, and they may belong to low-resource scenarios due to the lack of adequate data and prior knowledge. Therefore, studying knowledge graph completion methods in low-resource scenarios is crucial for solving practical problems. This paper proposes a low-resource knowledge graph completion method SMKGC (similarity messages knowledge graph completion), which combines nodes and edges message passing to predict links by sensing local relations and entities. Different with existing methods, SMKGC fuses the feature information of semantically similar nodes and semantically parallel edges with the Pathcon model, thereby to enhance the feature representation of message passing and improve link prediction accuracy. Specifically, it includes two modules: (1) Entity-similarity-based messaging, which aggregates information surrounding similar entities on the basis of capturing the neighboring edges of a given entity pair. (2) Relation-similarity-based messaging, which obtains the relative positions of a given entity pair through relational paths and combines the similar edges of the paths to perform relational prediction. Experimental results demonstrate that this method significantly outperforms other methods on benchmark datasets commonly used in knowledge graphs. The results also show that similar entities and relation messaging based on similar nodes can improve prediction accuracy in low-resource knowledge graph completion tasks.

Key words: knowledge graph, similarity, link prediction

张婷, 杜方, 宋丽娟, 史英杰, 赵国栋, 李婷. 结合实体和关系消息传递的低资源知识图谱补全[J]. 计算机工程与应用, 2024, 60(22): 137-144.

ZHANG Ting, DU Fang, SONG Lijuan, SHI Yingjie, ZHAO Guodong, LI Ting. Low-Resource Knowledge Graph Completion by Combining Entity and Relation Message Passing[J]. Computer Engineering and Applications, 2024, 60(22): 137-144.

参考文献

[1] WANG H, REN H, LESKOVEC J. Relational message passing for knowledge graph completion[C]//Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 2021: 1697-1707.
[2] BORDES A, USUNIER N, GARCIA-DURAN A, et al. Translating embeddings for modeling multi-relational data[C]//Advances in Neural Information Processing Systems 26, 2013.
[3] WANG Z, ZHANG J, FENG J, et al. Knowledge graph embedding by translating on hyperplanes[C]//Proceedings of the 28th AAAI Conference on Artificial Intelligence, 2014.
[4] KAOUDI Z, LORENZO A C M, MARKL V. Towards loosely-coupling knowledge graph embeddings and ontology-based reasoning[J]. arXiv:2202.03173, 2022.
[5] LU X, WANG L, JIANG Z, et al. MRE: a translational knowledge graph completion model based on multiple relation embedding[J]. Mathematical Biosciences and Engineering, 2023, 20: 5881-5900.
[6] WANG Q, MAO Z, WANG B, et al. GCN-Align: graph convolutional networks for aligning multi-relational knowledge graphs[C]//Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing, 2019: 3974-3984.
[7] SUN Z, DENG H, NIE Y, et al. Bootstrapping embeddings of knowledge graphs with entity descriptions for link prediction[C]//Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, 2018: 338-348.
[8] WANG Z, LV Q, LAN X, et al. Cross-lingual knowledge graph alignment via graph convolutional networks[C]//Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, 2018: 349-357.
[9] CHEN M, TIAN Y, YANG M, et al. MEDAL: multi-task embedded deep attributed learning for cross-lingual knowledge graph alignment[C]//Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, 2019: 2110-2120.
[10] FINN C, ABBEEL P, LEVINE S. Model-agnostic meta-learning for fast adaptation of deep networks[C]//Proceedings of the 34th International Conference on Machine Learning, 2017: 1126-1135.
[11] NICHOL A, ACHIAM J, SCHULMAN J, et al. On first-order meta-learning algorithms[C]//Proceedings of the 34th International Conference on Machine Learning, 2017: 2535-2544.
[12] CHEN M, ZHANG W, ZHANG W, et al. Meta relational learning for few-shot link prediction in knowledge graphs[J]. arXiv:1909.01515, 2019.
[13] 王子涵, 邵明光, 刘国军, 等. 基于实体相似度信息的知识图谱补全算法[J]. 计算机应用, 2018, 38(11): 3089-3093.
WANG Z H, SHAO M G, LIU G J, et al. Knowledge graph completion algorithm based on similarity between entities[J]. Journal of Computer Applications, 2018, 38(11): 3089-3093.
[14] ZHANG Y, LIANG H, JATOWT A, et al. GMH: a general multi-hop reasoning model for KG completion[J]. arXiv:2010.
07620, 2020.
[15] JI G, LIU K, HE S, et al. Knowledge graph completion with adaptive sparse transfer matrix[C]//Proceedings of the 30th AAAI Conference on Artificial Intelligence, 2016.
[16] AN Q, YU L. A heterogeneous network embedding framework for predicting similarity-based drug-target interactions[J]. Briefings in Bioinformatics, 2021, 22(6): bbab275.
[17] WANG Z H, LAI K, LI P J, et al. Tackling long-tailed relations and uncommon entities in knowledge graph completion[J]. arXiv:1909.11359, 2019.
[18] 张宁豫, 谢辛, 陈想, 等. 基于知识协同微调的低资源知识图谱补全方法[J]. 软件学报, 2022, 33(10): 3531-3545.
ZHANG N Y, XIE X, CHEN X, et al. Knowledge collaborative fine-tuning for low-resource knowledge graph completion[J]. Journal of Software, 2022, 33(10): 3531-3545.
[19] TROUILLON T, WELBL J, RIEDEL S, et al. Complex embeddings for simple link prediction[C]//Proceedings of the 33rd International Conference on Machine Learning, 2016: 2071-2080.
[20] YANG B, YIH W, HE X, et al. Embedding entities and relations for learning and inference in knowledge bases[J]. arXiv:1412.6575, 2014.
[21] SUN Z, DENG Z H, NIE J Y, et al. Rotate: knowledge graph embedding by relational rotation in complex space[J]. arXiv:1902.10197, 2019.
[22] KAZEMI S M, POOLE D. Simple embedding for link prediction in knowledge graphs[C]//Advances in Neural Information Processing Systems 31, 2018.
[23] ZHANG S, TAY Y, YAO L, et al. Quaternion knowledge graph embeddings[C]//Advances in Neural Information Processing Systems 32, 2019.
[24] SADEGHIAN A, ARMANDPOUR M, DING P, et al. DRUM: end-to-end differentiable rule mining on knowledge graphs[C]//Advances in Neural Information Processing Systems 32, 2019.