图书情报工作

图书情报工作 >

2021 , Vol. 65 >Issue 12: 101 - 111

DOI: https://doi.org/10.13266/j.issn.0252-3116.2021.12.010

研究论文

面向用户长短期偏好调节的可解释个性化推荐方法研究

  • 李伟卿 ,
  • 池毛毛 ,
  • 王伟军
展开
  • 1. 华中师范大学信息管理学院 武汉 430079;
    2. 华中师范大学青少年网络心理与行为教育部重点实验室 武汉 430079
李伟卿(ORCID:0000-0002-4108-2829),博士研究生;池毛毛(ORCID:0000-0003-2726-5933),副教授,博士。

收稿日期: 2020-12-09

  修回日期: 2021-02-24

  网络出版日期: 2021-07-03

基金资助

本文系国家自然科学基金项目"面向青少年网络适应的个性化信息服务优化方法研究"（项目编号：71974072）和国家自然科学基金项目"基于屏幕视觉热区的网络用户偏好提取及交互式个性化推荐研究"（项目编号：71571084）研究成果之一。

收起

Explainable Personalized Recommendation Method Based on Adjustment of Users' Long- and Short-Term Preferences

  • Li Weiqing ,
  • Chi Maomao ,
  • Wang Weijun
Expand
  • 1. School of Information Management, Central China Normal University, Wuhan 430079;
    2. Key Laboratory of Adolescent Cyber Psychology and Behavior, Ministry of Education, Wuhan 430079

Received date: 2020-12-09

  Revised date: 2021-02-24

  Online published: 2021-07-03

Fold

摘要

[目的/意义] 针对目前推荐模型愈加复杂、数据输入越来越多、传统推荐模型可解释性较低、推荐结果"过度特化"等问题，提出面向用户长短偏好调节的可解释个性化推荐方法。[方法/过程] 从用户近期产品需求及其长期生活方式两个维度构建用户长短偏好模型，借鉴用户评分偏置及注意力机制，将用户长短偏好与其评分相结合进行评分预测，从而形成Top-N推荐。[结果/结论] 通过在两个数据集上的实验结果表明，本方法对于不同的用户行为（显式反馈或隐式反馈），不同的推荐项目个数及在不同的推荐算法中都有良好表现。在无需对各种推荐模型进行较大改变的情况下，提升了推荐结果的准确率、召回率与多样性；另外基于对长短偏好系数的改变，实现对推荐结果多样性与准确率的调整，并且形成相应的推荐解释。

关键词： 个性化推荐; 用户偏好; 可解释性; 长短偏好调节

本文引用格式

李伟卿 , 池毛毛 , 王伟军 . 面向用户长短期偏好调节的可解释个性化推荐方法研究[J]. 图书情报工作, 2021 , 65(12) : 101 -111 . DOI: 10.13266/j.issn.0252-3116.2021.12.010

Abstract

[Purpose/significance] We put forward an explainable personalized recommendation method based on adjustment of users' long- and short-preferences in view of the current problems that increasingly complexity and more feature data inputs of recommendation models, low interpretability of traditional recommendation models and over-specialization of recommendation results.[Method/process] We constructed a user preference model from two dimensions of users' recent product needs and their long-term lifestyles, used the user's rating bias and attention mechanism for reference, combined the user's long- and short-term preference with their direct score to predict the score of unknown items, and formed the Top-N recommendations.[Result/conclusion] The experimental results on two datasets showed that our method had a good performance to different user behaviors (explicit feedback or implicit feedback), different number of Top-N recommended items, and in different recommendation algorithms. It improves the accuracy, recall and diversity of the recommendation results without making great changes to various recommendation models, and based on the change of long- and short-term preference coefficients, it realizes the adjustment to the diversity and accuracy of the recommendation results, and form the corresponding recommended explanation.

Key words： personalized recommendation; user preference; explainable recommendation; adjustment of long- and short-preference

参考文献

[1] LIU P, ZHANG L, GULLA J, et al. Dynamic attention-based explainable recommendation with textual and visual fusion[J]. Information processing & management, 2019, 57(6):102099.
[2] NAJI H. Improving recommendation system based on homophily principle and demographic[J]. Research journal of applied sciences, 2016, 11(10):1102-1106.
[3] ZHAO X, GUO Y, HE Y, et al. We know what you want to buy:a demographic-based system for product recommendation on microblogs[C]//ACM. KDD'14:proceedings of the 20th ACM SIGKDD international conference on knowledge discovery and data mining. New York:ACM press, 2014:1935-1944.
[4] HUANG Y, LIU H, LI W, et al. Lifestyles in amazon:evidence from online reviews enhanced recommender system[J]. International journal of market research, 2020, 62(6):689-706.
[5] 侯银秀, 李伟卿, 王伟军, 等. 基于用户偏好与商品属性情感匹配的图书个性化推荐研究[J].数据分析与知识发现, 2017, 1(8):9-17.
[6] BI J, LIU Y, FAN Z, et al. A deep neural networks based recommendation algorithm using user and item basic data[J]. International journal of machine learning and cybernetics, 2020, 11(4):763-777.
[7] 梁昌勇, 范汝鑫, 陆文星, 等. 基于CNN-LFM模型的个性化推荐[J].计算机仿真, 2020, 37(3):399-404.
[8] BEENEN G, LING K, WANG X, et al. Using social psychology to motivate contributions to online communities[C]//ACM. Proceedings of the 2004 ACM conference on computer supported cooperative work. New York:ACM press, 2004:212-221.
[9] BENNETT P. Modeling the impact of short-term and long-term behavior on search personalization[C]//ACM. International ACM SIGIR conference on research & development in information retrieval. New York:ACM, 2012:185-194.
[10] CHU Y, LEE J. The experiential preferences of the online consumers in different internet shopping lifestyles towards online shopping websites[C]//Springer. International Conference on Human-Computer Interaction. Lisbon:Springer, 2007.
[11] DIAZ A, GOMEZ M, MOLINA A, et al. A segmentation study of cinema consumers based on values and lifestyle[J]. Journal of retailing and consumer services, 2018, 3(41):79-89.
[12] SARKI I, BHUTTO N, ARSHAD I, et al. Impact of Pakistani university student's cultural values and lifestyles on meaning of brands[J]. Interdisciplinary journal of contemporary research in business, 2012, 3(9):643-654.
[13] SWINYARD W R. Shopping mall customer values:the national mall shopper and the list of values[J]. Journal of retailing and consumer services, 1998, 5(3):167-172.
[14] 潘煜, 高丽, 王方华. 生活方式、顾客感知价值对中国消费者购买行为影响[J].系统管理学报, 2009, 18(6):601-607.
[15] 潘煜, 罗莉娟, 刘丹, 等. 基于网络生活方式的个人用户云服务购买意愿研究[J]. 系统管理学报, 2013, 22(4):477-486.
[16] XIONG L, CHEN X, HUANG T, et al. Temporal collaborative filtering with Bayesian probabilistic tensor factorization[C]//SDM. Proceedings of the SIAM international conference on data mining. Philadelphia:SIAM, 2010:211-222.
[17] 张婷婷, 王伟军, 黄英辉, 等. 基于屏幕视觉热区的中文短文本关键词实时提取方法[J]. 情报学报, 2016, 35(12):1178-1188.
[18] 刘凯. 基于屏幕视觉热区的用户偏好提取及个性化推荐[M]. 北京:科学出版社, 2016.
[19] HU L, LI C, SHI C, et al. Graph neural news recommendation with long-term and short-term interest modeling[J]. Information processing and management, 2020, 57(2):1-10.
[20] CHEN D, ZHANG R, YUAN B, et al. Sequence-aware recommendation with long-term and short-term attention memory networks[C]//IEEE. 201920th IEEE international conference on mobile data management. IEEE, 2019:437-442.
[21] XIANG L, YUAN Q, ZHAO S, et al. Temporal recommendation on graphs via long- and short-term preference fusion[C]//ACM. KDD'10:proceedings of the 16th ACM SIGKDD international conference on knowledge discovery and data mining. New York:ACM press, 2010:723-732.
[22] 王微微, 夏秀峰, 李晓明. 一种基于用户行为反馈的兴趣度模型更新算法[J].辽宁大学学报(自然科学版), 2011, 38(1):40-45.
[23] GEDIKLI F, JANNACH D, GE M, et al. How should I explain? A comparison of different explanation types for recommender systems[J]. International journal of human-computer studies, 2014, 72(4):367-382.
[24] TINTAREV N, MASTHOFF J. Designing and evaluating explanations for recommender systems[M]. Boston:Recommender Systems Hand-book, Springer, 2011:479-510.
[25] ZHANG Y, CHEN X. Explainable recommendation:a survey and new perspectives[J]. ArXiv:information retrieval, 2018, 14(1):1-101.
[26] 余以胜, 韦锐, 刘鑫艳. 可解释的实时图书信息推荐模型研究[J].情报学报, 2019, 38(2):209-216.
[27] ZHANG Y, LAI G, ZHANG M, et al. Explicit factor models for explainable recommendation based on phrase-level sentiment analysis[C]//ACM. International ACM SIGIR conference on research and development in information retrieval. New York:ACM press, 2014:83-92.
[28] BAHDANAU D, CHO K, BENGIO Y. Neural machine translation by jointly learning to align and translate[EB/OL].[2021-04-01]. https://www.researchgate.net/publication/265252627_Neural_Machine_Translation_by_Jointly_Learning_to_Align_and_Translate.
[29] HU L, JIAN S, CAO L, et al. Interpretable recommendation via attraction modeling:learning multilevel attractiveness over multimodal movie contents[C]//AAAI. Twenty-seventh international joint conference on artificial intelligence IJCAI-18. Stockholm:AAAI Press, 2018, 7:3400-3406
[30] SEO S, HUANG J, YANG H, et al. Interpretable convolutional neural networks with dual local and global attention for review rating prediction[C]//ACM. RecSys'17:proceedings of the eleventh ACM conference on recommender systems. New York:ACM press, 2017, 8:297-305.
[31] YU J, ZHU H, CHANG C, et al. Influence function for unbiased recommendation[C]//ACM. SIGIR'20:the 43rd international ACM SIGIR conference on research and development in information retrieval. New York:ACM press, 2020, 7:1929-1932.
[32] 崔春生. 电子商务隐式浏览输入中的用户聚类分析[J].图书情报工作, 2011, 55(14):130-134.
[33] 王军, 张子柯. 基于社会化标签信息熵的个性化推荐算法[J].图书情报工作, 2013, 57(23):31-35.
[34] LIU H, HU Z, MIAN A, et al. A new user similarity model to improve the accuracy of collaborative filtering[J]. Knowledge-based systems, 2014, 56(1):156-166.
[35] DESHPANDE M, KARYPIS G. Item-based top-N recommendation algorithms[J]. ACM transactions on information systems, 2004, 22(1):143-177.
[36] POLAT H, DU W. SVD-based collaborative filtering with privacy[C]//ACM. SAC'05:proceedings of the 2005 ACM symposium on applied computing. New York:ACM press, 2005, 3:791-795.
[37] WU Y, LI Y, QIAN R. NE-UserCF:collaborative filtering recommender system model based on NMF and E2LSH[J]. International journal of performability engineering, 2017, 13(5):610-619.
[38] BI J W, LIU Y, FAN Z P. A deep neural networks based recommendation algorithm using user and item basic data[J]. International journal of machine learning and cybernetics, 2020, 11(4):763-777.
[39] DEVOOGHT R, BERSINI H. Long and short-term recommendations with recurrent neural networks[C]//ACM. UMAP'17:proceedings of the 25th conference on user modeling, adaptation and personalization. New York:ACM press, 2017, 7:13-21.
Options
文章导航

/

〈 〉