Signed-Graph Recommendation as Structural Consistency Maximization

2026-07-07

AI Summary

This work addresses representation bias in signed social recommendation caused by structural noise and data sparsity. The authors propose the SSC-Loop framework, which is the first to identify the lack of consistency in existing methods across structural, propagation, and semantic dimensions. To bridge the gap between noisy graph structures and reliable social semantics, the framework introduces an Enhanced Structural Alignment with Dual Augmentation (ESA-DA) module to strengthen structural consistency, designs a ternary propagation mechanism—distinguishing positive, negative, and neutral interactions—to ensure propagation consistency, and integrates contrastive learning to achieve semantic consistency. Experimental results demonstrate that the model significantly improves explicit rating prediction on Epinions and effectively leverages signed structural information for link existence prediction on Slashdot.

Abstract

While signed social recommendation has shown great potential by modeling both trust and distrust relations, its effectiveness is often hindered by structural noise and data sparsity. In this work, we first identify a fundamental inconsistency across the structural, propagation, and semantic layers of existing models, which leads to biased representations learned from sparse or noisy datasets. Furthermore, we observe that most existing methods treat the observed graph as fixed, failing to bridge the gap between noisy topologies and reliable social semantics. To address these issues, we propose a unified framework named SSC-Loop that treats signed social recommendation as the maximization of structural consistency. SSC-Loop includes three dedicated modules: ESA-DA for structural consistency, a P/N/O propagation mechanism for propagation consistency, and a contrastive learning objective for semantic consistency. Experiments on Epinions demonstrate that SSC-Loop achieves strong performance on explicit signed social rating prediction, while auxiliary results on Slashdot under a derived link-existence setting further suggest its ability to exploit signed social structures. Source code is available at https://github.com/Refrainwww/SSC-Loop.

Problem

• signed social recommendation
• structural noise
• data sparsity
• structural inconsistency
• social semantics

Innovation

• signed-graph recommendation
• structural consistency
• contrastive learning
• propagation mechanism
• social semantics