面向会话型推荐系统的个性化分层循环模型

doi:10.13190/j.jbupt.2019-143

北京邮电大学学报 ›› 2019, Vol. 42 ›› Issue (6): 142-148.doi: 10.13190/j.jbupt.2019-143

面向会话型推荐系统的个性化分层循环模型

王雅青, 郭彩丽, 楚云霏, 周洪弘, 冯春燕

1. 北京邮电大学信息与通信工程学院, 北京 100876;
2. 北京邮电大学先进信息网络北京实验室, 北京 100876

收稿日期:2019-07-09 出版日期:2019-12-28 发布日期:2019-11-15
通讯作者: 郭彩丽(1977-),女,教授,博士生导师,E-mail:guocaili@bupt.edu.cn. E-mail:guocaili@bupt.edu.cn
作者简介:王雅青(1990-),女,博士生.
基金资助:
国家重点研发计划项目（2018YFB1800805）

Personalized Hierarchical Recurrent Model for Session-Based Recommendation Systems

WANG Ya-qing, GUO Cai-li, CHU Yun-fei, ZHOU Hong-hong, FENG Chun-yan

1. School of Information and Communication Engineering, Beijing University of Posts and Telecommunications, Beijing 100876, China;
2. Beijing Laboratory of Advanced Information Networks, Beijing University of Posts and Telecommunications, Beijing 100876, China

Received:2019-07-09 Online:2019-12-28 Published:2019-11-15

摘要/Abstract

摘要： 为了精准地捕捉用户行为模式，引入中期兴趣的概念，提出一个基于循环神经网络（RNN）的个性化分层循环模型，通过在同一框架下联合利用用户的会话、区块和全部行为序列来学习用户的综合兴趣.利用一个捕捉会话内序列模式的会话级RNN建模用户的短期兴趣；设计了一个捕捉区块内相邻会话关联关系的区块级RNN，进一步描述用户的中期兴趣；使用一个用户级RNN追踪长期兴趣的演化；引入带有不同交互机制的融合层，以有效融合不同层次的兴趣信息.在3个真实数据集上进行实验，结果表明，该方法与先进的推荐方法相比，Recall@10提升了18.35%.

关键词: 会话型推荐系统, 循环神经网络, 个性化推荐

Abstract: The existing studies of session-based recommendations mainly focus on the short-term and long-term interests of users. In order to accurately depict behavior patterns of users, the author introduces the medium-term interests and proposes personalized hierarchical recurrent model (PHRM) based on recurrent neural networks (RNNs), to learn a comprehensive description of user interests by jointly leveraging session, block and global behaviors in a unified framework. First, to model short-term interests, a session-level RNN is designed to capture sequential patterns in sessions. Next, to further describe medium-term interests, a block-level RNN is added to capture correlations across sessions in a block. Then, a user-level RNN is devised to track evolution of long-term interests. Finally, the article designs fusion layers with different interaction mechanisms to effectively integrate cross-level interest information. Simulations on three real-world datasets show that PHRM outperforms the state-of-the-art recommendation methods, with Recall@10 increasing by 18.35%.

Key words: session-based recommendation systems, recurrent neural networks, personalized recommendations

中图分类号:

TP301

王雅青, 郭彩丽, 楚云霏, 周洪弘, 冯春燕. 面向会话型推荐系统的个性化分层循环模型[J]. 北京邮电大学学报, 2019, 42(6): 142-148.

WANG Ya-qing, GUO Cai-li, CHU Yun-fei, ZHOU Hong-hong, FENG Chun-yan. Personalized Hierarchical Recurrent Model for Session-Based Recommendation Systems[J]. JOURNAL OF BEIJING UNIVERSITY OF POSTS AND TELECOM, 2019, 42(6): 142-148.

参考文献

[1] Ludewig M, Jannach D. Evaluation of session-based recommendation algorithms[J]. User Modeling and User-Adapted Interaction, 2018, 28(4-5):331-390.
[2] Barkan O, Koenigstein N. Item2vec:neural item embedding for collaborative filtering[C]//MLSP. Salerno:IEEE, 2016:1-6.
[3] Hidasi B, Karatzoglou A, Baltrunas L, et al. Session-based recommendations with recurrent neural networks[J]. arXiv preprint arXiv:1511. 06939, 2015.
[4] Hidasi B, Karatzoglou A. Recurrent neural networks with top-k gains for session-based recommendations[C]//Conference on Information and Knowledge Management (CIKM). Turin:ACM, 2018:843-852.
[5] Zhao W, Wang B, Ye J, et al. Plastic:prioritize long and short-term information in top-n recommendation using adversarial training[C]//IJCAI. Stockholm:IJCAI Organization, 2018:3676-3682.
[6] Quadrana M, Karatzoglou A, Hidasi B, et al. Personalizing session-based recommendations with hierarchical recurrent neural networks[C]//Proceedings of the 11^th ACM Conference on Recommender Systems-RecSys'17. New York:ACM Press, 2017:130-137.
[7] Sarwar B M, Karypis G, Konstan J, et al. Item-based collaborative filtering recommendation algorithms[C]//Proceedings of the 10^th international conference on World Wide Web-WWW'01. New York:ACM Press, 2001:285-295.
[8] Jannach D, Ludewig M. When recurrent neural networks meet the neighborhood for session-based recommendation[C]//Proceedings of the 11^th ACM Conference on Recommender Systems-RecSys'17. New York:ACM Press, 2017:306-310.
[9] Cho K, Van Merriënboer B, Gulcehre C, et al. Learning phrase representations using RNN encoder-decoder for statistical machine translation[J]. arXiv preprint arXiv:1406. 1078, 2014.
[10] Qu Yanru, Cai Han, Ren Kan, et al. Product-based neural networks for user response prediction[C]//2016 IEEE 16^th International Conference on Data Mining (ICDM). Piscataway:IEEE Press, 2016:1149-1154.
[11] Rendle S, Freudenthaler C, Gantner Z, et al. Bpr:bayesian personalized ranking from implicit feedback[C]//UAI. Montreal:AUAI Press, 2009:452-461.

面向会话型推荐系统的个性化分层循环模型

Personalized Hierarchical Recurrent Model for Session-Based Recommendation Systems

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 8

编辑推荐

Metrics

本文评价

[1]	李剑, 陈宇航. 一种多视角高精度图片的深度估计方法[J]. 北京邮电大学学报, 2021, 44(5): 101-106.
[2]	蒲悦逸, 王文涵, 朱强, 陈朋朋. 基于CNN-ResNet-LSTM模型的城市短时交通流量预测算法[J]. 北京邮电大学学报, 2020, 43(5): 9-14.
[3]	袁仁进, 陈刚, 李锋, 魏双建. 基于VSM和Bisecting K-means聚类的新闻推荐方法[J]. 北京邮电大学学报, 2019, 42(1): 114-119.
[4]	苏放, 王晓宇, 张治. 基于注意力机制的评论摘要生成[J]. 北京邮电大学学报, 2018, 41(3): 7-13.
[5]	王晓军, 付超. 利用模糊分块改进协同过滤的扩展性和准确性[J]. 北京邮电大学学报, 2017, 40(1): 74-78.
[6]	王晓军. 推荐系统中分布式混合协同过滤方法[J]. 北京邮电大学学报, 2016, 39(2): 25-29.
[7]	王鑫, 黄忠义. 网络资源中基于K-Means聚类的个性化推荐[J]. 北京邮电大学学报, 2014, 37(s1): 120-124.
[8]	王晓军. 利用项目属性和偏好改进协同过滤推荐[J]. 北京邮电大学学报, 2014, 37(6): 68-71,76.