Social Recommendation

Social recommendation aims to improve personalized recommendations by leveraging social connections among users, assuming that connected users share similar preferences. Current research focuses on addressing challenges like data sparsity and noise in social networks, employing techniques such as graph neural networks, diffusion models, and contrastive learning to refine user representations and mitigate biases like popularity bias. These advancements enhance recommendation accuracy and address privacy concerns through methods like federated learning, impacting both the development of more effective recommendation systems and the understanding of social influence on user behavior.