MHGCN: Multiview highway graph convolutional network for cross-lingual entity alignment

• DOI: 10.26599/tst.2021.9010056
• Publication Date: 2021-12-09T21:09:15Z
• Authors: Jianliang Gao; Xiangyue Liu; Yibo Chen; Fan Xiong
• Keywords: Knowledge graph; Entity linking; Similarity (geometry)

Knowledge graphs (KGs) provide a wealth of prior knowledge for the research on social networks. Cross-lingual entity alignment aims at integrating complementary KGs from different languages and thus benefits various knowledge-driven network studies. Recent methods often take an embedding-based approach to model relation embedding KGs. However, these studies mostly focus information itself its structural features but ignore influence multiple types data in In this paper, we propose new framework named multiview highway graph convolutional (MHGCN), which considers views semantic, attribute. To learn entity, MHGCN employs (GCN) each view. addition, weights fuses according importance view obtain better embedding. The entities are identified based similarity embeddings. experimental results show that consistently outperforms state-of-the-art methods. also will benefit fusion through cross-lingual KG alignment.