arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2108.04607cs.IR
收藏

用于推荐的全双曲图卷积网络

Fully Hyperbolic Graph Convolution Network for Recommendation

Liping Wang, Fenyu Hu, Shu Wu, Liang Wang

摘要:

最近,基于图卷积网络(GCN)的方法已经取得了出色的性能推荐。这些方法将用户和项目嵌入到欧氏空间中,并在用户-项目交互图上执行图卷积。然而,现实世界中的数据集通常表现出树状层次结构,这使得欧氏空间在捕捉用户-项目关系时效率较低。相比之下,双曲空间,作为一个连续的模拟树图,提供了一个有前途的选择。在本文中,我们提出了一个全双曲GCN模型的建议,其中所有的操作都在双曲空间中进行。利用双曲空间的优势,我们的方法是能够嵌入用户/项目失真较小,更准确地捕捉用户与项目的交互关系。在公共基准数据集上进行的大量实验表明,我们的方法优于欧几里得和双曲型方法,并且需要更低的嵌入维数才能达到相当的性能。

英文摘要:

Recently, Graph Convolution Network (GCN) based methods have achieved outstanding performance for recommendation. These methods embed users and items in Euclidean space, and perform graph convolution on user-item interaction graphs. However, real-world datasets usually exhibit tree-like hierarchical structures, which make Euclidean space less effective in capturing user-item relationship. In contrast, hyperbolic space, as a continuous analogue of a tree-graph, provides a promising alternative. In this paper, we propose a fully hyperbolic GCN model for recommendation, where all operations are performed in hyperbolic space. Utilizing the advantage of hyperbolic space, our method is able to embed users/items with less distortion and capture user-item interaction relationship more accurately. Extensive experiments on public benchmark datasets show that our method outperforms both Euclidean and hyperbolic counterparts and requires far lower embedding dimensionality to achieve comparable performance.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑