MDSEA: Knowledge Graph Entity Alignment Based on Multimodal Data Supervision

被引:1
作者
Fang, Jianyong [1 ,2 ]
Yan, Xuefeng [1 ]
机构
[1] Nanjing Univ Aeronaut & Astronaut, Coll Comp Sci & Technol, Nanjing 210000, Peoples R China
[2] Jiangsu Automat Res Inst, Lianyungang 222000, Peoples R China
来源
APPLIED SCIENCES-BASEL | 2024年 / 14卷 / 09期
基金
中国国家自然科学基金;
关键词
multimodal data supervision; entity alignment; knowledge graph; transformer;
D O I
10.3390/app14093648
中图分类号
O6 [化学];
学科分类号
0703 ;
摘要
With the development of social media, the internet, and sensing technologies, multimodal data are becoming increasingly common. Integrating these data into knowledge graphs can help models to better understand and utilize these rich sources of information. The basic idea of the existing methods for entity alignment in knowledge graphs is to extract different data features, such as structure, text, attributes, images, etc., and then fuse these different modal features. The entity similarity in different knowledge graphs is calculated based on the fused features. However, the structures, attribute information, image information, text descriptions, etc., of different knowledge graphs often have significant differences. Directly integrating different modal information can easily introduce noise, thus affecting the effectiveness of the entity alignment. To address the above issues, this paper proposes a knowledge graph entity alignment method based on multimodal data supervision. First, Transformer is used to obtain encoded representations of knowledge graph entities. Then, a multimodal supervised method is used for learning the entity representations in the knowledge graph so that the vector representations of the entities contain rich multimodal semantic information, thereby enhancing the generalization ability of the learned entity representations. Finally, the information from different modalities is mapped to a shared low-dimensional subspace, making similar entities closer in the subspace, thus optimizing the entity alignment effect. The experiments on the DBP15K dataset compared with methods such as MTransE, JAPE, EVA, DNCN, etc., all achieve optimal results.
引用
收藏
页数:12
相关论文
共 37 条
[1]  
Bordes A., 2013, Adv. Neural Inf. Process. Syst., V26
[2]  
Cao YX, 2019, 57TH ANNUAL MEETING OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS (ACL 2019), P1452
[3]   Multi-modal Siamese Network for Entity Alignment [J].
Chen, Liyi ;
Li, Zhi ;
Xu, Tong ;
Wu, Han ;
Wang, Zhefeng ;
Yuan, Nicholas Jing ;
Chen, Enhong .
PROCEEDINGS OF THE 28TH ACM SIGKDD CONFERENCE ON KNOWLEDGE DISCOVERY AND DATA MINING, KDD 2022, 2022, :118-126
[4]   MMEA: Entity Alignment for Multi-modal Knowledge Graph [J].
Chen, Liyi ;
Li, Zhi ;
Wang, Yijun ;
Xu, Tong ;
Wang, Zhefeng ;
Chen, Enhong .
KNOWLEDGE SCIENCE, ENGINEERING AND MANAGEMENT (KSEM 2020), PT I, 2020, 12274 :134-147
[5]  
Chen MH, 2021, 16TH CONFERENCE OF THE EUROPEAN CHAPTER OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS (EACL 2021), P645
[6]  
Chen MH, 2017, PROCEEDINGS OF THE TWENTY-SIXTH INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE, P1511
[7]  
Guo LB, 2019, PR MACH LEARN RES, V97
[8]   FedCKE: Cross-Domain Knowledge Graph Embedding in Federated Learning [J].
Huang, Wei ;
Liu, Jia ;
Li, Tianrui ;
Ji, Shenggong ;
Wang, Dexian ;
Huang, Tianqiang .
IEEE TRANSACTIONS ON BIG DATA, 2023, 9 (03) :792-804
[9]  
Kipf T.N., 2017, INT C LEARN REPR, P1
[10]   A Novel Embedding Model for Knowledge Graph Entity Alignment Based on Graph Neural Networks [J].
Li, Hongchan ;
Han, Zhaoyang ;
Zhu, Haodong ;
Qian, Yuchao .
APPLIED SCIENCES-BASEL, 2023, 13 (10)