UTCS: Effective Unsupervised Temporal Community Search with Pre-training of Temporal Dynamics and Subgraph Knowledge

TL;DR

This paper introduces UTCS, an unsupervised method for temporal community search that leverages pre-training of temporal dynamics and subgraph knowledge, overcoming limitations of traditional and learning-based approaches.

Contribution

The paper proposes a novel unsupervised framework with pre-training for temporal community search, addressing the challenge of unknown subgraph structures and capturing temporal interactions.

Findings

Outperforms existing methods on five real-world datasets

Effectively captures temporal interaction information

Does not require predefined subgraph structures

Abstract

In many real-world applications, the evolving relationships between entities can be modeled as temporal graphs, where each edge has a timestamp representing the interaction time. As a fundamental problem in graph analysis, {\it community search (CS)} in temporal graphs has received growing attention but exhibits two major limitations: (1) Traditional methods typically require predefined subgraph structures, which are not always known in advance. (2) Learning-based methods struggle to capture temporal interaction information. To fill this research gap, in this paper, we propose an effective \textbf{U}nsupervised \textbf{T}emporal \textbf{C}ommunity \textbf{S}earch with pre-training of temporal dynamics and subgraph knowledge model (\textbf{\model}). \model~contains two key stages: offline pre-training and online search. In the first stage, we introduce multiple learning objectives to facilitate the pre-training process in the unsupervised learning setting. In the second stage, we identify a candidate subgraph and compute community scores using the pre-trained node representations and a novel scoring mechanism to determine the final community members. Experiments on five real-world datasets demonstrate the effectiveness.