Link Prediction on Textual Edge Graphs

Chen Ling, Zhuofeng Li, Yuntong Hu, Zhongyuan Liu, Zheng Zhang, Shuang Zheng, Jian Pei, Liang Zhao

Keywords: Link Prediction, Large Language Model, Graph Neural Network

Abstract: Textual-edge Graphs (TEGs), characterized by rich text annotations on edges, are increasingly significant in network science due to their ability to capture rich contextual information among entities. Existing works have proposed various edge-aware graph neural networks (GNNs) or let language models directly make predictions. However, they often fail to fully capture the contextualized semantics on edges and graph topology, respectively. This inadequacy is particularly evident in link prediction tasks that require a comprehensive understanding of graph topology and semantics between nodes. In this paper, we present a novel framework - \textsc{Link2Doc}, designed especially for link prediction on TEGs. Specifically, we propose to summarize neighborhood information between node pairs as a human-written document to preserve both semantic and topology information. We also present a specialized GNN framework to process the multi-scaled interaction between target nodes in a stratified manner. Finally, a self-supervised learning model is utilized to enhance the GNN's text-understanding ability from language models. Empirical evaluations, including link prediction, edge classification, parameter analysis, runtime comparison, and ablation studies, on five real-world datasets demonstrate that \textsc{Link2Doc} achieves generally better performance against existing edge-aware GNNs and language models in link predictions.

Primary Area: learning on graphs and other geometries & topologies

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Submission Number: 8284