Learning graph attention-aware knowledge graph embedding
文献类型：期刊论文
| 作者 | Li, Chen1; Peng, Xutan6; Niu, Yuhang1,2; Zhang, Shanghang3; Peng, Hao1,5; Zhou, Chuan4 ; Li, Jianxin1
|
| 刊名 | NEUROCOMPUTING
 |
| 出版日期 | 2021-10-21 |
| 卷号 | 461页码:516-529 |
| 关键词 | Knowledge graph embedding Graph attention mechanism Entity typing Link prediction |
| ISSN号 | 0925-2312 |
| DOI | 10.1016/j.neucom.2021.01.139 |
| 英文摘要 | The knowledge graph, which utilizes graph structure to represent multi-relational data, has been widely used in the reasoning and prediction tasks, attracting considerable research efforts recently. However, most existing works still concentrate on learning knowledge graph embeddings straightforwardly and intuitively without subtly considering the context of knowledge. Specifically, recent models deal with each single triple independently or consider contexts indiscriminately, which is one-sided as each knowledge unit (i.e., triple) can be derived from its partial surrounding triples. In this paper, we propose a graph-attention-based model to encode entities, which formulates a knowledge graph as an irregular graph and explores a number of concrete and interpretable knowledge compositions by integrating the graph-structured information via multiple independent channels. To measure the correlation between entities from different angles (i.e., entity pair, relation, and structure), we respectively develop three attention metrics. By making use of our enhanced entity embeddings, we further introduce several improved factorization functions for updating relation embeddings and evaluating candidate triples. We conduct extensive experiments on downstream tasks including entity classification, entity typing, and link prediction to validate our methods. Empirical results validate the importance of our introduced attention metrics and demonstrate that our proposed method can improve the performance of factorization models on large-scale knowledge graphs. (c) 2021 Elsevier B.V. All rights reserved. |
| 资助项目 | Natural Science Foundation of China program[U20B2053] ; Natural Science Foundation of China program[62002007] ; Natural Science Foundation of China program[61772151] ; S&T Program of Hebei[20310101D] ; S&T Program of Hebei[SKLSDE-2020ZX-12] |
| WOS研究方向 | Computer Science |
| 语种 | 英语 |
| WOS记录号 | WOS:000704086300003 |
| 出版者 | ELSEVIER |
| 源URL | [http://ir.amss.ac.cn/handle/2S8OKBNM/59347]  |
| 专题 | 应用数学研究所 |
| 通讯作者 | Li, Chen |
| 作者单位 | 1.Beihang Univ, Beijing Adv Innovat Ctr Big Data & Brain Comp, Beijing, Peoples R China 2.Beihang Univ, State Key Lab Software Dev Environm Lab, Beijing, Peoples R China 3.Carnegie Mellon Univ, Dept Elect & Comp Engn, Pittsburgh, PA USA 4.Chinese Acad Sci, Acad Math & Syst Sci, Beijing, Peoples R China 5.Beihang Univ, Sch Cyber Sci & Technol, Beijing, Peoples R China 6.Univ Sheffield, Dept Comp Sci, Sheffield, S Yorkshire, England |
| 推荐引用方式 GB/T 7714 | Li, Chen,Peng, Xutan,Niu, Yuhang,et al. Learning graph attention-aware knowledge graph embedding[J]. NEUROCOMPUTING,2021,461:516-529. |
| APA | Li, Chen.,Peng, Xutan.,Niu, Yuhang.,Zhang, Shanghang.,Peng, Hao.,...&Li, Jianxin.(2021).Learning graph attention-aware knowledge graph embedding.NEUROCOMPUTING,461,516-529. |
| MLA | Li, Chen,et al."Learning graph attention-aware knowledge graph embedding".NEUROCOMPUTING 461(2021):516-529. |
入库方式： OAI收割
来源：数学与系统科学研究院
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。