SPAN: Subgraph Prediction Attention Network for Dynamic Graphs

TL;DR

This paper introduces SPAN, a novel attention-based model for predicting the evolution of subgraphs in dynamic graphs, outperforming existing methods in accuracy across various real-world datasets.

Contribution

The paper presents a new end-to-end subgraph prediction model with a cross-attention twin-tower mechanism for dynamic graphs, improving prediction accuracy over state-of-the-art methods.

Findings

Outperforms existing models with a 5.02% to 10.88% accuracy gain.

Effective in both homogeneous and heterogeneous dynamic graphs.

Demonstrates robustness across multiple real-world datasets.

Abstract

This paper proposes a novel model for predicting subgraphs in dynamic graphs, an extension of traditional link prediction. This proposed end-to-end model learns a mapping from the subgraph structures in the current snapshot to the subgraph structures in the next snapshot directly, i.e., edge existence among multiple nodes in the subgraph. A new mechanism named cross-attention with a twin-tower module is designed to integrate node attribute information and topology information collaboratively for learning subgraph evolution. We compare our model with several state-of-the-art methods for subgraph prediction and subgraph pattern prediction in multiple real-world homogeneous and heterogeneous dynamic graphs, respectively. Experimental results demonstrate that our model outperforms other models in these two tasks, with a gain increase from 5.02% to 10.88%.