A subgraph interpretation generative model for knowledge graph link prediction based on uni-relation transformation

Cong Yuan; Junping Yao; Xiaojun Li; Hao Wang; Yi Guo

doi:10.1117/12.3009388

16 October 2023 A subgraph interpretation generative model for knowledge graph link prediction based on uni-relation transformation

Cong Yuan, Junping Yao, Xiaojun Li, Hao Wang, Yi Guo

Author Affiliations +

Proceedings Volume 12803, Fifth International Conference on Artificial Intelligence and Computer Science (AICS 2023); 1280339 (2023) https://doi.org/10.1117/12.3009388
Event: 2023 5th International Conference on Artificial Intelligence and Computer Science (AICS 2023), 2023, Wuhan, China

Abstract

In recent years, Graph Neural Networks (GNNs) have witnessed rapid development. Their strengths in capturing topological information of graph data contribute to significant performance improvements in tasks such as knowledge graph (KG) link prediction. To understand the reason for this performance improvement, it is necessary to extract the subgraph patterns learned by GNNs from KGs. Nevertheless, the accuracy of existing GNN interpreters has not been validated in explaining multi-relation graph data, such as KGs, and related tools have not been implemented yet, leading to difficulties in extracting explanation subgraphs. To address this problem, this paper proposes a KG link prediction model that converts multi-relation KGs into uni-relation graphs. This model combines entities in the KG into new nodes, and treats relations as features of the new nodes, thereby creating a graph with only a single relation. A denoising autoencoder is then trained on the new graph for link prediction, and a GNN interpreter is used to generate subgraph explanations. Experiments on three benchmark datasets show that the proposed model based on uni-relation graph transformation significantly enhances the relative AUC, as compared to GraIL without transformation. Finally, an explanation subgraph extraction experiment is performed on the FB15K-237 dataset, demonstrating the effectiveness of the model in directly extracting link predictions for explanation.

(2023) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Cong Yuan, Junping Yao, Xiaojun Li, Hao Wang, and Yi Guo "A subgraph interpretation generative model for knowledge graph link prediction based on uni-relation transformation", Proc. SPIE 12803, Fifth International Conference on Artificial Intelligence and Computer Science (AICS 2023), 1280339 (16 October 2023); https://doi.org/10.1117/12.3009388

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
11 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Education and training

Data modeling

Performance modeling

Denoising

Neural networks

Reflection

Connectors

Show All Keywords

Keywords/Phrases

Search In:

Publication Years