Learning Social Graph for Inactive User Recommendation

TL;DR

This paper introduces LSIR, a novel social recommendation method that learns an optimized social graph to improve recommendations for inactive users by refining social connections and mimicking active user behaviors.

Contribution

The paper proposes a new approach called LSIR that learns an optimal social graph and employs mimic learning to enhance inactive user recommendations.

Findings

Achieves up to 129.58% improvement in NDCG for inactive users

Effectively refines social graphs by removing noisy edges and adding meaningful ones

Demonstrates significant performance gains on real-world datasets

Abstract

Social relations have been widely incorporated into recommender systems to alleviate data sparsity problem. However, raw social relations don't always benefit recommendation due to their inferior quality and insufficient quantity, especially for inactive users, whose interacted items are limited. In this paper, we propose a novel social recommendation method called LSIR (\textbf{L}earning \textbf{S}ocial Graph for \textbf{I}nactive User \textbf{R}ecommendation) that learns an optimal social graph structure for social recommendation, especially for inactive users. LSIR recursively aggregates user and item embeddings to collaboratively encode item and user features. Then, graph structure learning (GSL) is employed to refine the raw user-user social graph, by removing noisy edges and adding new edges based on the enhanced embeddings. Meanwhile, mimic learning is implemented to guide active users in mimicking inactive users during model training, which improves the construction of new edges for inactive users. Extensive experiments on real-world datasets demonstrate that LSIR achieves significant improvements of up to 129.58\% on NDCG in inactive user recommendation. Our code is available at~\url{https://github.com/liun-online/LSIR}.