HSR: Hyperbolic Social Recommender

With the prevalence of online social media, users' social connections have been widely studied and utilized to enhance the performance of recommender systems. In this paper, we explore the use of hyperbolic geometry for social recommendation. We present the Hyperbolic Social Recommender (HSR), a novel social recommendation framework that utilizes hyperbolic geometry to boost the performance. With the help of hyperbolic space, HSR can learn high-quality user and item representations to better model user-item interaction and user-user social relations. Through extensive experiments on four real-world datasets, we show that our proposed HSR outperforms its Euclidean counterpart and state-of-the-art social recommenders in click-through rate prediction and top -K recommendation, demonstrating the effectiveness of social recommendation in the hyperbolic space. (c) 2021 Elsevier Inc. All rights reserved.

Automated summary

This paper explores the use of hyperbolic geometry in social recommendation, introducing the Hyperbolic Social Recommender (HSR) which can learn high-quality user and item representations in hyperbolic space to better model user-item interaction and user-user social relations. Extensive experiments show that HSR outperforms its Euclidean counterpart and state-of-the-art social recommenders in click-through rate prediction and top-K recommendation, demonstrating the effectiveness of social recommendation in the hyperbolic space.