高级搜索

留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

基于行为延迟共享网络的个性化商品推荐方法

张红霞 董燕辉 肖军弼 杨勇进

张红霞, 董燕辉, 肖军弼, 杨勇进. 基于行为延迟共享网络的个性化商品推荐方法[J]. 电子与信息学报, 2021, 43(10): 2993-3000. doi: 10.11999/JEIT200964
引用本文: 张红霞, 董燕辉, 肖军弼, 杨勇进. 基于行为延迟共享网络的个性化商品推荐方法[J]. 电子与信息学报, 2021, 43(10): 2993-3000. doi: 10.11999/JEIT200964
Hongxia ZHANG, Yanhui DONG, Junbi XIAO, Yongjin YANG. Personalized Commodity Recommendation Method Based on Behavioral Delay Sharing Network[J]. Journal of Electronics & Information Technology, 2021, 43(10): 2993-3000. doi: 10.11999/JEIT200964
Citation: Hongxia ZHANG, Yanhui DONG, Junbi XIAO, Yongjin YANG. Personalized Commodity Recommendation Method Based on Behavioral Delay Sharing Network[J]. Journal of Electronics & Information Technology, 2021, 43(10): 2993-3000. doi: 10.11999/JEIT200964

基于行为延迟共享网络的个性化商品推荐方法

doi: 10.11999/JEIT200964
基金项目: 国家重点研发计划(2018YFC1406204),国家自然科学基金(61872385),中央高校基本科研业务费专项资金(18CX02140A)
详细信息
    作者简介:

    张红霞:女,1981年生,博士,副教授,研究方向为服务计算,边缘计算,机器学习

    董燕辉:男,1996年生,硕士生,研究方向为服务推荐,机器学习

    肖军弼:男,1968年生,硕士,副教授,研究方向为计算机网络,软件定义网络

    杨勇进:男,1996年生,硕士生,研究方向为边缘计算,深度学习

    通讯作者:

    张红霞 zhanghx@upc.edu.cn

  • 中图分类号: TP391

Personalized Commodity Recommendation Method Based on Behavioral Delay Sharing Network

Funds: The National Key Research and Development Project(2018YFC1406204), The National Natural Science Foundation of China (61872385), The Fundamental Research Funds for the Central University (18CX02140A)
  • 摘要: 针对电商平台难以利用历史浏览行为进行个性化商品推荐的问题,该文提出了一种行为延迟共享网络模型(BDSN),充分结合历史浏览信息,对用户进行精准浏览推荐。该模型提出行为延迟门控循环神经单元(BDGRU),将历史浏览时间间隔作为用户活跃度因子,对神经元状态进行更新,用于计算用户的兴趣表示。为了提高向量表示的一致性,该模型提出共享参数网络,将用户侧和商品侧的表示向量收敛到统一空间,解决个性化商品推荐点击率预估问题。并在真实数据集上进行实验,结果表明,BDSN模型在验证集上的AUC指标和损失函数均处于最优,在测试集上的AUC指标相较基本模型提高37%,能够有效提升商品推荐的准确性。
  • 图  1  行为延迟门控循环单元结构图

    图  2  共享参数向量表示网络结构

    图  3  BDSN模型结构

    图  4  商品推荐流程图

    图  5  BDSN与其它对比方法的验证集精确度指标比

    图  6  BDSN与其它对比方法的训练损失值变化情况

    图  7  行为延迟门控单元对状态值的影响

    图  8  共享网络对用户和商品嵌入表示的影响

    图  9  BDSN用户和商品侧的向量可视化

    表  1  实验数据集统计信息

    数据集用户数量商品数量样本数
    训练集67891089162435690
    验证集16641328022450
    测试集16641567021216
    下载: 导出CSV

    表  2  不同方法模型在测试集上的AUC值

    模型AUCRelaImpr(%)
    BDSN0.873837
    DREN0.845427
    DNN(BaseModel)0.77120
    WideDeep0.838224
    PNN0.840425
    GRUSN0.825820
    BD-Sep0.806412
    下载: 导出CSV
  • [1] 岳金果. 考虑消费者感知与网络效应的平台推送决策研究[D]. [硕士论文], 西南财经大学, 2019. doi: 10.27412/d.cnki.gxncu.2019.001880.

    YUE Jinguo. Research on platform recommendation under consumer perception and network effect[D]. [Master dissertation], Southwestern University of Finance and Economics, 2019. doi: 10.27412/d.cnki.gxncu.2019.001880.
    [2] JANNACH D and JUGOVAC M. Measuring the business value of recommender systems[J]. ACM Transactions on Management Information Systems (TMIS) , 2019, 10(4): 16. doi: 10.1145/3370082
    [3] 吕刚, 张伟. 基于深度学习的推荐系统应用综述[J]. 软件工程, 2020, 23(2): 5–8. doi: 10.19644/j.cnki.issn2096-1472.2020.02.002

    LV Gang and ZHANG Wei. Survey of deep learning applied in recommendation system[J]. Software Engineering, 2020, 23(2): 5–8. doi: 10.19644/j.cnki.issn2096-1472.2020.02.002
    [4] NABIZADEH A H, LEAL J P, RAFSANJANI H N, et al. Learning path personalization and recommendation methods: A survey of the state-of-the-art[J]. Expert Systems with Applications, 2020, 159: 113596. doi: 10.1016/j.eswa.2020.113596
    [5] HYUN D, PARK C, CHO J, et al. Learning to utilize auxiliary reviews for recommendation[J]. Information Sciences, 2021, 545: 595–607. doi: 10.1016/j.ins.2020.09.025
    [6] CHENG H T, KOC L, HARMSEN J, et al. Wide & deep learning for recommender systems[C]. The 1st Workshop on Deep Learning for Recommender Systems, Boston, USA, 2016: 7–10. doi: 10.1145/2988450.2988454.
    [7] COVINGTON P, ADAMS J, and SARGIN E. Deep neural networks for YouTube recommendations[C]. The 10th ACM Conference on Recommender Systems, Boston, USA, 2016: 191–198. doi: 10.1145/2959100.2959190.
    [8] ZHANG Jiang, WANG Yufeng, YUAN Zhiyuan, et al. Personalized real-time movie recommendation system: Practical prototype and evaluation[J]. Tsinghua Science and Technology, 2020, 25(2): 180–191. doi: 10.26599/TST.2018.9010118
    [9] QU Yanru, FANG Bohui, ZHANG Weinan, et al. Product-based neural networks for user response prediction over multi-field categorical data[J]. ACM Transactions on Information Systems (TOIS) , 2019, 37(1): 5. doi: 10.1145/3233770
    [10] 王娜, 何晓明, 刘志强, 等. 一种基于用户播放行为序列的个性化视频推荐策略[J]. 计算机学报, 2020, 43(1): 123–135. doi: 10.11897/SP.J.1016.2020.00123

    WANG Na, HE Xiaoming, LIU Zhiqiang, et al. Personalized video recommendation strategy based on user's playback behavior sequence[J]. Chinese Journal of Computers, 2020, 43(1): 123–135. doi: 10.11897/SP.J.1016.2020.00123
    [11] 夏永生, 王晓蕊, 白鹏, 等. 基于时序和距离的门控循环单元兴趣点推荐算法[J]. 计算机工程, 2020, 46(1): 52–59. doi: 10.19678/j.issn.1000-3428.0053659

    XIA Yongsheng, WANG Xiaorui, BAI Peng, et al. Point of interest recommendation algorithm of gated recurrent unit based on time series and distance[J]. Computer Engineering, 2020, 46(1): 52–59. doi: 10.19678/j.issn.1000-3428.0053659
    [12] 李宇琦, 陈维政, 闫宏飞, 等. 基于网络表示学习的个性化商品推荐[J]. 计算机学报, 2019, 42(8): 1767–1778. doi: 10.11897/SP.J.1016.2019.01767

    LI Yuqi, CHEN Weizheng, YAN Hongfei, et al. Learning graph-based embedding for personalized product recommendation[J]. Chinese Journal of Computers, 2019, 42(8): 1767–1778. doi: 10.11897/SP.J.1016.2019.01767
    [13] DHELIM S, AUNG N, and NING Huansheng. Mining user interest based on personality-aware hybrid filtering in social networks[J]. Knowledge-Based Systems, 2020, 206: 106227. doi: 10.1016/j.knosys.2020.106227
    [14] ZHOU Guorui, ZHU Xiaoqiang, SONG Chenru, et al. Deep interest network for click-through rate prediction[C]. The 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, London, UK, 2018: 1059–1068. doi: 10.1145/3219819.3219823.
    [15] ZHOU Guorui, MOU Na, FAN Ying, et al. Deep interest evolution network for click-through rate prediction[C]. The AAAI Conference on Artificial Intelligence, Honolulu, USA, 2019: 5941–5948. doi: 10.1609/aaai.v33i01.33015941.
    [16] ZHENG Guanjie, ZHANG Fuzheng, ZHENG Zihan, et al. DRN: A deep reinforcement learning framework for news recommendation[C]. Proceedings of the 2018 World Wide Web Conference, Lyon, France, 2018. doi: 10.1145/3178876.3185994.
    [17] 曲朝阳, 宋晨晨, 任有学, 等. 结合用户活跃度的协同过滤推荐算法[J]. 东北电力大学学报, 2017, 37(5): 74–79. doi: 10.19718/j.issn.1005-2992.2017.05.015

    QU Zhaoyang, SONG Chenchen, REN Youxue, et al. Recommendations based on collaborative filtering by user activity[J]. Journal of Northeast Electric Power University, 2017, 37(5): 74–79. doi: 10.19718/j.issn.1005-2992.2017.05.015
    [18] 王锦坤, 姜元春, 孙见山, 等. 考虑用户活跃度和项目流行度的基于项目最近邻的协同过滤算法[J]. 计算机科学, 2016, 43(12): 158–162. doi: 10.11896/j.issn.1002-137X.2016.12.028

    WANG Jinkun, JIANG Yuanchun, SUN Jianshan, et al. A collaborative filtering algorithm based on item nearest neighbors considering user activity and item popularity[J]. Computer Science, 2016, 43(12): 158–162. doi: 10.11896/j.issn.1002-137X.2016.12.028
    [19] 司亚利, 张付志, 刘文远. 基于签到活跃度和时空概率模型的自适应兴趣点推荐方法[J]. 电子与信息学报, 2020, 42(3): 678–686. doi: 10.11999/JEIT190287

    SI Yali, ZHANG Fuzhi, and LIU Wenyuan. An adaptive point-of-interest recommendation method based on check-in activity and temporal-spatial probabilistic models[J]. Journal of Electronics &Information Technology, 2020, 42(3): 678–686. doi: 10.11999/JEIT190287
    [20] 于帅, 林宣雄, 邱媛媛. 大规模隐式反馈的词向量音乐推荐模型[J]. 计算机系统应用, 2017, 26(11): 28–35. doi: 10.15888/j.cnki.csa.006049

    YU Shuai, LIN Xuanxiong, and QIU Yuanyuan. Implicit music recommender based on large scale word-embedding[J]. Computer Systems &Applications, 2017, 26(11): 28–35. doi: 10.15888/j.cnki.csa.006049
    [21] 赵晨阳, 王俊岭. 基于隐含上下文支持向量机的服务推荐方法[J]. 通信学报, 2019, 40(9): 61–73. doi: 10.11959/j.issn.1000-436x.2019190

    ZHAO Chenyang and WANG Junling. Service recommendation method based on context-embedded support vector machine[J]. Journal on Communications, 2019, 40(9): 61–73. doi: 10.11959/j.issn.1000-436x.2019190
    [22] 潘建林, 汪彬, 董晓晨. 基于SICAS消费者行为模型的社交电商模式及比较研究[J]. 企业经济, 2020, 39(10): 37–43. doi: 10.13529/j.cnki.enterprise.economy.2020.10.005

    PAN Jianlin, WANG Bin, and DONG Xiaochen. Social e-commerce model and comparative research based on SICAS consumer behavior model[J]. Enterprise Economy, 2020, 39(10): 37–43. doi: 10.13529/j.cnki.enterprise.economy.2020.10.005
    [23] 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. doi: 10.3772/j.issn.1000-0135.2019.02.010

    YU Yisheng, WEI Rui, and LIU Xinyan. Explainable real-time book information recommendation model[J]. Journal of the China Society for Scientific and Technical Information, 2019, 38(2): 209–216. doi: 10.3772/j.issn.1000-0135.2019.02.010
    [24] TIANCHI. Inc. KDD Cup 2020 challenges for modern E-commerce platform: Debiasing[EB/OL]. https://tianchi.aliyun.com/specials/promotion/kdd2020-cn, 2020.
  • 加载中
图(9) / 表(2)
计量
  • 文章访问数:  773
  • HTML全文浏览量:  247
  • PDF下载量:  71
  • 被引次数: 0
出版历程
  • 收稿日期:  2020-11-09
  • 修回日期:  2021-06-28
  • 网络出版日期:  2021-08-09
  • 刊出日期:  2021-10-18

目录

    /

    返回文章
    返回