Abstract:The rise of online social networks has facilitated the evolution of social recommender systems, which incorporate social relations to enhance users' decision-making process. With the great success of Graph Neural Networks (GNNs) in learning node representations, GNN-based social recommendations have been widely studied to model user-item interactions and user-user social relations simultaneously. Despite their great successes, recent studies have shown that these advanced recommender systems are highly vulnerable to adversarial attacks, in which attackers can inject well-designed fake user profiles to disrupt recommendation performances. While most existing studies mainly focus on argeted attacks to promote target items on vanilla recommender systems, untargeted attacks to degrade the overall prediction performance are less explored on social recommendations under a black-box scenario. To perform untargeted attacks on social recommender systems, attackers can construct malicious social relationships for fake users to enhance the attack performance. However, the coordination of social relations and item profiles is challenging for attacking black-box social recommendations. To address this limitation, we first conduct several preliminary studies to demonstrate the effectiveness of cross-community connections and cold-start items in degrading recommendations performance. Specifically, we propose a novel framework MultiAttack based on multi-agent reinforcement learning to coordinate the generation of cold-start item profiles and cross-community social relations for conducting untargeted attacks on black-box social recommendations. Comprehensive experiments on various real-world datasets demonstrate the effectiveness of our proposed attacking framework under the black-box setting.

Research Fronts of Robust Social Recommendation

A Novel Social Recommendation Method Fusing User’s Social Status and Homophily Based on Matrix Factorization Techniques

Robustness Analysis of Triangle Relations Attack in Social Recommender Systems

Exploiting Structural and Temporal Influence for Dynamic Social-Aware Recommendation

Personalized Recommender Systems Based on Social Relationships and Historical Behaviors

A Multi-theoretical Framework for Social Network-based Recommendation.

Robust Federated Recommendation System

Advanced Privacy-Preserving Federated Relationship Recommendation

Multi-agent Attacks for Black-box Social Recommendations

A comprehensive social matrix factorization for recommendations with prediction and feedback mechanisms by fusing trust relationships and social tags

Interactive Social Recommendation

Social recommendation across multiple relational domains.

A Synthetic Approach for Recommendation: Combining Ratings, Social Relations, and Reviews

Social Recommendation With Learning Personal and Social Latent Factors

An enhanced social matrix factorization model for recommendation based on social networks using social interaction factors

A Secure And Efficient Framework For Privacy Preserving Social Recommendation

Recommender System Based on Social Trust Relationships

Towards Robust Recommendation: A Review and an Adversarial Robustness Evaluation Library

Social Contextual Recommendation

Research on Collaborative Filtering Recommendation Based on Trust Relationship and Rating Trust

Personalized recommendation system based on social relationships and historical behaviors