CoDCL: Counterfactual-Inspired Augmentation Contrastive Learning for Temporal Link Prediction in Social Networks

TL;DR

This paper introduces CoDCL, a novel framework that enhances temporal link prediction in social networks by integrating counterfactual-inspired data augmentation with contrastive learning, improving adaptability to evolving structures.

Contribution

The paper presents a universal plug-and-play module combining counterfactual augmentation and contrastive learning for improved dynamic network representation.

Findings

CoDCL significantly outperforms existing methods on multiple datasets.

The framework effectively captures temporal changes in interaction patterns.

It can be integrated into various models without architectural changes.

Abstract

Temporal link prediction is crucial for rapidly growing social networks. Existing methods often overlook the underlying causal mechanisms that drive link formation, making it difficult for algorithms to adapt to complex structures that continuously evolve over time. To enable prediction models to adapt to complex temporal environments, they need to be robust to emerging structural changes. We propose a dynamic network learning framework CoDCL, which combines counterfactual-inspired augmentation with contrastive learning to address this deficiency. Furthermore, we devise a comprehensive strategy to generate high-quality counterfactual data, combining a dynamic treatments design with efficient structural neighborhood exploration to quantify the temporal changes in interaction patterns. Crucially, the entire CoDCL is designed as a plug-and-play universal module that can be seamlessly integrated into various existing temporal graph models without requiring architectural modifications. Extensive experiments conducted on multiple real-world datasets demonstrate that CoDCL significantly outperforms state-of-the-art baselines in temporal link prediction, highlighting the effectiveness of integrating counterfactual-inspired data augmentation into dynamic representation learning.