Fully Hyperbolic Graph Convolution Network for Recommendation | |
Wang,Liping; Hu,Fenyu; Wu,Shu; Wang,Liang | |
2021 | |
会议名称 | CIKM 2021 |
会议日期 | November 1–5, 2021 |
会议地点 | Virtual Event, Australia |
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. |
语种 | 英语 |
七大方向——子方向分类 | 数据挖掘 |
国重实验室规划方向分类 | 其他 |
是否有论文关联数据集需要存交 | 否 |
文献类型 | 会议论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/52178 |
专题 | 智能感知与计算研究中心 |
通讯作者 | Wu,Shu |
作者单位 | Center for Research on Intelligent Perception and Computing, Institute of Automation, Chinese Academy of Sciences |
第一作者单位 | 中国科学院自动化研究所 |
通讯作者单位 | 中国科学院自动化研究所 |
推荐引用方式 GB/T 7714 | Wang,Liping,Hu,Fenyu,Wu,Shu,et al. Fully Hyperbolic Graph Convolution Network for Recommendation[C],2021. |
条目包含的文件 | ||||||
文件名称/大小 | 文献类型 | 版本类型 | 开放类型 | 使用许可 | ||
3459637.3482109.pdf(1734KB) | 会议论文 | 开放获取 | CC BY-NC-SA | 浏览 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论