Rating Distribution Calibration for Selection Bias Mitigation in Recommendations

被引:9
|
作者
Liu, Haochen [1 ]
Tang, Da [2 ]
Yang, Ji [2 ]
Zhao, Xiangyu [3 ]
Liu, Hui [1 ]
Tang, Jiliang [1 ]
Cheng, Youlong [2 ]
机构
[1] Michigan State Univ, E Lansing, MI 48824 USA
[2] ByteDance Inc, Beijing, Peoples R China
[3] City Univ Hong Kong, Hong Kong, Peoples R China
来源
PROCEEDINGS OF THE ACM WEB CONFERENCE 2022 (WWW'22) | 2022年
基金
美国国家科学基金会;
关键词
recommendation system; unbiased recommendation; self-supervised learning;
D O I
10.1145/3485447.3512078
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
Real-world recommendation datasets have been shown to be subject to selection bias, which can challenge recommendation models to learn real preferences of users, so as to make accurate recommendations. Existing approaches to mitigate selection bias, such as data imputation and inverse propensity score, are sensitive to the quality of the additional imputation or propensity estimation models. To break these limitations, in this work, we propose a novel selfsupervised learning (SSL) framework, i.e., Rating Distribution Calibration (RDC), to tackle selection bias without introducing additional models. In addition to the original training objective, we introduce a rating distribution calibration loss. It aims to correct the predicted rating distribution of biased users by taking advantage of that of their similar unbiased users. We empirically evaluate RDC on two real-world datasets and one synthetic dataset. The experimental results show that RDC outperforms the original model as well as the state-of-the-art debiasing approaches by a significant margin.
引用
收藏
页码:2048 / 2057
页数:10
相关论文
共 50 条
  • [21] Soft calibration for selection bias problems under mixed-effects models
    Gao, Chenyin
    Yang, Shu
    Kim, Jae Kwang
    BIOMETRIKA, 2023, 110 (04) : 897 - 911
  • [22] Rating Bias and Preference Acquisition
    Freyne, Jill
    Berkovsky, Shlomo
    Smith, Gregory
    ACM TRANSACTIONS ON INTERACTIVE INTELLIGENT SYSTEMS, 2013, 3 (03)
  • [23] Selection bias mitigation in recommender system using uninteresting items based on temporal visibility
    Shi, Lei
    Li, Shuqing
    Ding, Xiaowei
    Bu, Zhan
    EXPERT SYSTEMS WITH APPLICATIONS, 2023, 213
  • [24] Your Best Guess When You Know Nothing: Identification and Mitigation of Selection Bias
    Dost, Katharina
    Taskova, Katerina
    Riddle, Patricia
    Wicker, Jorg
    20TH IEEE INTERNATIONAL CONFERENCE ON DATA MINING (ICDM 2020), 2020, : 996 - 1001
  • [25] Selective rating: partisan bias in crowdsourced news rating systems
    Duncan, Megan
    JOURNAL OF INFORMATION TECHNOLOGY & POLITICS, 2022, 19 (03) : 360 - 375
  • [26] Calibration with low bias
    Withers, Christopher S.
    Nadarajah, Saralees
    STATISTICAL PAPERS, 2013, 54 (02) : 371 - 379
  • [27] Calibration with low bias
    Christopher S. Withers
    Saralees Nadarajah
    Statistical Papers, 2013, 54 : 371 - 379
  • [28] Load Distribution and VFD Topology Selection for Harmonic Mitigation in an Optimal Way
    Thakur, Pratapaditya
    IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, 2020, 56 (01) : 48 - 56
  • [29] Asymmetric Power Device Rating Selection for Even Temperature Distribution in NPC Inverter
    Choi, Ui-Min
    Blaabjerg, Frede
    2017 IEEE ENERGY CONVERSION CONGRESS AND EXPOSITION (ECCE), 2017, : 4196 - 4201
  • [30] Small sample and selection bias effects in calibration under latent factor regression models
    Sundberg, Rolf
    JOURNAL OF CHEMOMETRICS, 2007, 21 (5-6) : 227 - 238