Details

Autor(en) / Beteiligte

• Liu, Xiao
• Mian, Li
• Dong, Yuxiao
• Zhang, Fanjin
• Zhang, Jing
• Tang, Jie
• Zhang, Peng
• Gong, Jibing
• Wang, Kuansan

Titel

OAG } know : Self-Supervised Learning for Linking Knowledge Graphs

Ist Teil von

• IEEE transactions on knowledge and data engineering, 2023-02, Vol.35 (2), p.1895-1908

Ort / Verlag

IEEE

Erscheinungsjahr

2023

Quelle

IEEE Xplore

Beschreibungen/Notizen

• We propose a self-supervised embedding learning framework-SelfLinKG-to link concepts in heterogeneous knowledge graphs. Without any labeled data, SelfLinKG can achieve competitive performance against its supervised counterpart, and significantly outperforms state-of-the-art unsupervised methods by 26%-50% under linear classification protocol. The essential components of SelfLinKG are local attention-based encoding and momentum contrastive learning. The former aims to learn the graph representation using an attention network, while the latter is to learn a self-supervised model across knowledge graphs using contrastive learning. SelfLinKG has been deployed to build the the new version, called OAG<inline-formula><tex-math notation="LaTeX">_{\mathrm {know}}</tex-math> <mml:math><mml:msub><mml:mrow/><mml:mi> know </mml:mi></mml:msub></mml:math><inline-graphic xlink:href="liu-ieq2-3090830.gif"/> </inline-formula> of Open Academic Graph (OAG). All data and codes are publicly available.