Explaining Synergistic Effects in Social Recommendations

TL;DR

This work tackles the explainability gap in social recommendation caused by nonlinear synergistic effects across multi-view networks. It introduces SemExplainer, an information-theoretic explainer that defines synergistic subgraphs via mutual-information criteria and refines them through conditional-entropy optimization, finally producing readable path explanations. Empirical results on ACM, Last-FM, and AugCitation show SemExplainer achieves superior SIS and SIN scores while maintaining fidelity, demonstrating clearer insights into how cross-view signals jointly drive recommendations. The approach advances trust and interpretability in complex multi-view social recommender systems and lays groundwork for broader global explainability of synergistic information.

Abstract

In social recommenders, the inherent nonlinearity and opacity of synergistic effects across multiple social networks hinders users from understanding how diverse information is leveraged for recommendations, consequently diminishing explainability. However, existing explainers can only identify the topological information in social networks that significantly influences recommendations, failing to further explain the synergistic effects among this information. Inspired by existing findings that synergistic effects enhance mutual information between inputs and predictions to generate information gain, we extend this discovery to graph data. We quantify graph information gain to identify subgraphs embodying synergistic effects. Based on the theoretical insights, we propose SemExplainer, which explains synergistic effects by identifying subgraphs that embody them. SemExplainer first extracts explanatory subgraphs from multi-view social networks to generate preliminary importance explanations for recommendations. A conditional entropy optimization strategy to maximize information gain is developed, thereby further identifying subgraphs that embody synergistic effects from explanatory subgraphs. Finally, SemExplainer searches for paths from users to recommended items within the synergistic subgraphs to generate explanations for the recommendations. Extensive experiments on three datasets demonstrate the superiority of SemExplainer over baseline methods, providing superior explanations of synergistic effects.

Figures (8)

• Figure 1: The illustration of synergistic effects. The system synergistically integrates information from three views: (1) User A and User B are friends; (2) User B listens to Music D; and (3) Music D and Music G both belong to the rock genre. Through the interaction of these views, the system infers that User A is likely to enjoy Music G.
• Figure 2: Overall structure of SemExplainer.
• Figure 3: Ablation analysis across three datasets. ‘w/o ESE’ denotes removal of the explanatory subgraph extraction component, ‘w/o SID’ denotes removal of the synergistic information disentanglement component.
• Figure 4: The effect of hyperparameter $\lambda$ on metrics FID+, FID-, SIS, SIN, and SPA.
• Figure 5: Questionnaire results for the three datasets. The blue box plot depicts the score distribution for Case A, whereas the red box plot depicts the distribution for Case B. The red dot denotes the mean, and the two black lines indicate the maximum and minimum values.