A semi-supervised framework fusing multiple information for knowledge graph entity alignment

Open Webpage

Zepeng Li, Nengneng Ding, Chenhui Liang, Shuo Cao, Minyu Zhai, Rikui Huang, Zhenwen Zhang, Bin Hu

Published: 01 Jan 2025, Last Modified: 27 Jul 2025Expert Syst. Appl. 2025EveryoneRevisionsBibTeXCC BY-SA 4.0
Abstract: Highlights•Propose a novel semi-supervised framework for the entity alignment task.•Introduce the HGNN in the entity alignment task to collect graph structural information.•Refine the pseudo label selection process with the similarity matrices.•Enhance the entity representation from a more comprehensive perspective.