| 摘要: |
| 实体对齐(entity alignment, EA)旨在寻找不同知识图谱(knowledge graph, KG)中等价实体. 目前, 基于嵌入的EA方法存在以下局限性. 首先, KG中的异构结构没有完全建模. 其次, 文本信息的使用受限于词嵌入. 第三, 对齐推理算法缺乏探索. 针对上述限制, 提出基于异构图注意力网络的实体对齐方法(heterogeneous graph attention network for entity alignment, HGAT-EA). HGAT-EA包括两个通道, 一个用于学习结构嵌入, 另一个用于学习字符级语义嵌入. 第1个通道采用异构图注意力网络(heterogeneous graph attention network, HGAT). HGAT充分利用了异构结构和关系三元组来学习实体嵌入. 第2个通道是利用字符级字面量来学习字符级语义嵌入. HGAT-EA通过多通道考虑多个视图, 并通过HGAT充分利用异构结构. HGAT-EA考虑了3种不同的对齐推理算法. 实验结果证明了该方法的有效性, 进一步结合实验结果对HGAT-EA的不同组件进行详细分析, 并给出相应的结论. |
| 关键词: 实体对齐 知识图谱融合 异构图注意力网络 表示学习 数据集成 |
| DOI：10.13328/j.cnki.jos.007371 |
| 分类号:TP18 |
| 基金项目:国家自然科学基金(62002262, 62172082, 62072086, 62072084) |
|
| Heterogeneous Graph Attention Network for Entity Alignment |
|
SUN Chen-Chen1,2, JIN Yu-Yuan1,2, SHEN De-Rong3, NIE Tie-Zheng3, KOU Yue3
|
|
1.Chang’an University, Xi’an 710018, China;2.Tianjin University of Technology, Tianjin 300384, China;3.Northeastern University, Shenyang 110189, China
|
| Abstract: |
| Entity alignment (EA) aims to identify equivalent entities across different knowledge graph (KG). Embedding-based EA methods still have several limitations, listed below. First, the heterogeneous structures within KGs are not fully modeled. Second, the utilization of text information is constrained by word embeddings. Third, alignment inference algorithms are underexplored. To address these limitations, this study proposes a heterogeneous graph attention network for entity alignment (HGAT-EA). HGAT-EA consists of two channels: one for learning structural embeddings and the other for learning character-level semantic embeddings. The first channel employs a heterogeneous graph attention network (HGAT), which fully leverages heterogeneous structures and relation triples to learn entity embeddings. The second channel utilizes character-level literals to learn character-level semantic embeddings. HGAT-EA incorporates multiple views through these channels and maximizes the use of heterogeneous structures through HGAT. HGAT-EA introduces three alignment inference algorithms. Experimental results validate the effectiveness of HGAT-EA. Following these results, we provide detailed analyses of the various components of HGAT-EA and present the corresponding conclusions. |
| Key words: entity alignment (EA) knowledge graph (KG) fusion heterogeneous graph attention network (HGAT) representation learning data integration |
