Summary
International Symposium on Nonlinear Theory and Its Applications
2022
Session Number:B3L-E
Session:
Number:B3L-E-02
Common Space Learning with Gaussian Embedding for Multi-Modal Entity Alignment
Kenta Hama , Takashi Matsubara,
pp.355-358
Publication Date:12/12/2022
Online ISSN:2188-5079
DOI:10.34385/proc.71.B3L-E-02
PDF download (317.6KB)
Summary:
Knowledge graphs are used in various systems that represent relationships between objects as a graph. Entity alignment is the task of finding entities that have the same object between two knowledge graphs. When utilizing multi-modal information for entity alignment, it is important to neglect unnecessary information contained in the data set. However, existing methods such as MMEA fail to consider the importance of information for each modal. In this study, we propose a method that expresses the importance of each information as the probability distribution. The proposed method outperforms MMEA in the entity alignment task of two multi-modal knowledge graphs.