Journal of Frontiers of Computer Science and Technology ›› 2023, Vol. 17 ›› Issue (10): 2343-2357.DOI: 10.3778/j.issn.1673-9418.2307053

• Special Issue on Large Language Models and Knowledge Graphs • Previous Articles     Next Articles

Contrast Research of Representation Learning in Entity Alignment Based on Graph Neural Network

PENG Huang, ZENG Weixin, ZHOU Jie, TANG Jiuyang, ZHAO Xiang   

  1. Laboratory for Big Data and Decision, National University of Defense Technology, Changsha 410073, China
  • Online:2023-10-01 Published:2023-10-01

基于图神经网络的实体对齐表示学习方法比较研究

彭鐄,曾维新,周杰,唐九阳,赵翔   

  1. 国防科技大学 大数据与决策实验室,长沙 410073

Abstract: Entity alignment is an important step in knowledge fusion, which aims to identify equivalent entities in different knowledge graphs. In order to accurately determine the equivalent entities, the existing methods first perform representation learning to map the entities into a low-dimensional vector space, and then infer the equivalence of the entities by the similarity between the vectors. Recent works on entity alignment focus on the improvement of representation learning methods. In order to better understand the mechanism of these models, mine valuable design directions, and provide reference for subsequent optimization and improvement work, this paper reviews the research on representation learning methods for entity alignment. Firstly, based on the existing methods, a general framework for representation learning is proposed, and several representative works are summarized and analyzed. Then, these works are compared and analyzed through experiments, and the common methods of each module in the framework are compared. Through the results, the advantages and disadvantages of various methods are summarized, and the use suggestions are put forward. Finally, the feasibility of the alignment and fusion of large language models and knowledge graphs is preliminarily discussed, and the existing problems and challenges are analyzed.

Key words: knowledge fusion, entity alignment, representation learning, graph neural network, large language model

摘要: 实体对齐是知识融合的一个重要步骤,其目的在于识别不同知识图谱中的等价实体。为准确判断出对等的实体,现有方法首先进行表示学习,将实体映射到低维向量空间中,接着通过向量间的相似度推断实体的等价性。而近期实体对齐的相关工作也大都聚焦于表示学习方法的改进上。为了能够更好地理解这些模型的机理,挖掘有价值的设计思路,并为后续的优化改进工作提供参考,对实体对齐表示学习方法进行了研究综述。首先基于现有方法,提出了一个通用的表示学习框架,并用该框架对几个具有代表性的工作进行了归纳概括以及分析解构。接着通过实验对这些工作进行了对比分析,并对框架中各个模块的常见方法进行了比较。根据实验结果,总结了各种方法的优劣,并提出了使用建议。最后初步讨论了大规模语言模型与知识图谱对齐融合的可行性,并分析了存在的问题以及潜在的挑战。

关键词: 知识融合, 实体对齐, 表示学习, 图神经网络, 语言大模型