Dual Metric Learning for Effective and Efficient Cross-Domain Recommendations

被引:26
作者
Li, Pan [1 ]
Tuzhilin, Alexander [1 ]
机构
[1] NYU, Stern Sch Business, Dept Technol Operat & Stat, New York, NY 10012 USA
基金
澳大利亚研究理事会;
关键词
Cross domain recommendation; dual learning; metric learning; orthogonal mapping; ALGORITHMS;
D O I
10.1109/TKDE.2021.3074395
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Cross domain recommender systems have been increasingly valuable for helping consumers identify useful items in different applications. However, existing cross-domain models typically require large number of overlap users, which can be difficult to obtain in some applications. In addition, they did not consider the duality structure of cross-domain recommendation tasks, thus failing to take into account bidirectional latent relations between users and items and achieve optimal recommendation performance. To address these issues, in this paper we propose a novel cross-domain recommendation model based on dual learning that transfers information between two related domains in an iterative manner until the learning process stabilizes. We develop a novel latent orthogonal mapping to extract user preferences over multiple domains while preserving relations between users across different latent spaces. Furthermore, we combine the dual learning method with the metric learning approach, which allows us to significantly reduce the required common user overlap across the two domains and leads to even better cross-domain recommendation performance. We test the proposed model on two large-scale industrial datasets and six domain pairs, demonstrating that it consistently and significantly outperforms all the state-of-the-art baselines. We also show that the proposed model works well with very few overlap users to obtain satisfying recommendation performance comparable to the state-of-the-art baselines that use many overlap users.
引用
收藏
页码:321 / 334
页数:14
相关论文
共 78 条
  • [21] Ioffe Sergey, 2015, International conference on machine learning, V37, P448, DOI DOI 10.48550/ARXIV.1502.03167
  • [22] Kanagawa Heishiro, 2019, Advances in Information Retrieval. 41st European Conference on IR Research, ECIR 2019. Proceedings: Lecture Notes in Computer Science (LNCS 11438), P20, DOI 10.1007/978-3-030-15719-7_3
  • [23] Kedem D., 2012, P ADV NEUR INF PROC, P2573
  • [24] Khoshneshin M., 2010, P 4 ACM C RECOMMENDE, P87
  • [25] Kingma D. P., 2014, P 2 INT C LEARN REPR
  • [26] Transfer Learning via Contextual Invariants for One-to-Many Cross-Domain Recommendation
    Krishnan, Adit
    Das, Mahashweta
    Bendre, Mangesh
    Yang, Hao
    Sundaram, Hari
    [J]. PROCEEDINGS OF THE 43RD INTERNATIONAL ACM SIGIR CONFERENCE ON RESEARCH AND DEVELOPMENT IN INFORMATION RETRIEVAL (SIGIR '20), 2020, : 1081 - 1090
  • [27] Metric Learning: A Survey
    Kulis, Brian
    [J]. FOUNDATIONS AND TRENDS IN MACHINE LEARNING, 2013, 5 (04): : 287 - 364
  • [28] Lee DD, 2001, ADV NEUR IN, V13, P556
  • [29] Li B, 2009, 21ST INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE (IJCAI-09), PROCEEDINGS, P2052
  • [30] PURS: Personalized Unexpected Recommender System for Improving User Satisfaction
    Li, Pan
    Que, Maofei
    Jiang, Zhichao
    Hu, Yao
    Tuzhilin, Alexander
    [J]. RECSYS 2020: 14TH ACM CONFERENCE ON RECOMMENDER SYSTEMS, 2020, : 279 - 288