Fully Hyperbolic Graph Convolution Network for Recommendation

被引:8
|
作者
Wang, Liping [1 ,2 ]
Hu, Fenyu [1 ,2 ]
Wu, Shu [1 ,2 ,3 ]
Wang, Liang [1 ,2 ]
机构
[1] Chinese Acad Sci, Ctr Res Intelligent Percept & Comp, Inst Automat, Beijing, Peoples R China
[2] Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing, Peoples R China
[3] Chinese Acad Sci, Artificial Intelligence Res, Beijing, Peoples R China
基金
中国国家自然科学基金;
关键词
recommendation system; graph neural networks; hyperbolic space;
D O I
10.1145/3459637.3482109
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
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.
引用
收藏
页码:3483 / 3487
页数:5
相关论文
共 50 条
  • [1] Adaptive Boosting Graph Convolution Network for Recommendation
    Guo, Tong
    Wu, Bin
    Zhang, Xue
    [J]. 2021 IEEE INTERNATIONAL CONFERENCE ON INFORMATION COMMUNICATION AND SOFTWARE ENGINEERING (ICICSE 2021), 2021, : 122 - 125
  • [2] Enhancing Graph Convolution Network for Novel Recommendation
    Ma, Xuan
    Qian, Tieyun
    Liang, Yile
    Sun, Ke
    Yun, Hang
    Zhang, Mi
    [J]. DATABASE SYSTEMS FOR ADVANCED APPLICATIONS, DASFAA 2022, PT II, 2022, : 69 - 84
  • [3] SocialLGN: Light graph convolution network for social recommendation
    Liao, Jie
    Zhou, Wei
    Luo, Fengji
    Wen, Junhao
    Gao, Min
    Li, Xiuhua
    Zeng, Jun
    [J]. INFORMATION SCIENCES, 2022, 589 : 595 - 607
  • [4] Quaternion-based graph convolution network for recommendation
    Fang, Yaxing
    Zhao, Pengpeng
    Liu, Guanfeng
    Liu, Yanchi
    Sheng, Victor S. S.
    Zhao, Lei
    Zhou, Xiaofang
    [J]. WORLD WIDE WEB-INTERNET AND WEB INFORMATION SYSTEMS, 2023, 26 (05): : 2835 - 2854
  • [5] Recommendation System Based on Perceptron and Graph Convolution Network
    Lian, Zuozheng
    Yin, Yongchao
    Wang, Haizhen
    [J]. CMC-COMPUTERS MATERIALS & CONTINUA, 2024, 79 (03): : 3939 - 3954
  • [6] Quaternion-based graph convolution network for recommendation
    Yaxing Fang
    Pengpeng Zhao
    Guanfeng Liu
    Yanchi Liu
    Victor S. Sheng
    Lei Zhao
    Xiaofang Zhou
    [J]. World Wide Web, 2023, 26 : 2835 - 2854
  • [7] LightGCN: Simplifying and Powering Graph Convolution Network for Recommendation
    He, Xiangnan
    Deng, Kuan
    Wang, Xiang
    Li, Yan
    Zhang, Yongdong
    Wang, Meng
    [J]. PROCEEDINGS OF THE 43RD INTERNATIONAL ACM SIGIR CONFERENCE ON RESEARCH AND DEVELOPMENT IN INFORMATION RETRIEVAL (SIGIR '20), 2020, : 639 - 648
  • [8] Graph convolution neural network for recommendation using graph negative sampling
    Huang, Heyuan
    Mu, Caihong
    Fang, Yunfei
    Liu, Yi
    [J]. Xi'an Dianzi Keji Daxue Xuebao/Journal of Xidian University, 2024, 51 (01): : 86 - 99
  • [9] Hyperbolic multichannel graph convolutional network for social recommendation
    Yang, Xingyao
    Chang, Mengxue
    Yu, Jiong
    Wang, Dongxiao
    Dang, Zibo
    [J]. Journal of Intelligent and Fuzzy Systems, 2024, 46 (04): : 9543 - 9557
  • [10] Recommendation Algorithm Based on Deep Light Graph Convolution Network in Knowledge Graph
    Chen, Xiaobin
    Xiao, Nanfeng
    [J]. ADVANCES IN INFORMATION RETRIEVAL, ECIR 2023, PT I, 2023, 13980 : 216 - 231