基于用户会话的TF-Ranking推荐方法

doi:10.3969/j.issn.0255-8297.2021.03.014

应用科学学报 ›› 2021, Vol. 39 ›› Issue (3): 495-494.doi: 10.3969/j.issn.0255-8297.2021.03.014

• 计算机科学与应用 • 上一篇

基于用户会话的TF-Ranking推荐方法

贾丹¹, 孙静宇²

1. 太原理工大学信息与计算机学院, 山西太原 030024;
2. 太原理工大学软件学院, 山西太原 030024

收稿日期:2020-05-08 发布日期:2021-06-08
通信作者: 孙静宇，副教授，研究方向为协同Web搜索、推荐系统、智慧城市。E-mail:1065036845@qq.com E-mail:1065036845@qq.com
基金资助:
山西省科技厅重点研发计划项目基金（No.201803D31226）；山西省研究生教育创新项目基金（No.2019SY117）资助

TF-Ranking Recommendation Method Based on User Session

JIA Dan¹, SUN Jingyu²

1. College of Information and Computer, Taiyuan University of Technology, Taiyuan 030024, Shanxi, China;
2. College of Software, Taiyuan University of Technology, Taiyuan 030024, Shanxi, China

Received:2020-05-08 Published:2021-06-08

摘要/Abstract

摘要： 基于用户会话日志，提出了一种融合XGBoost和门控循环单元网络的TF-Ranking推荐方法。该方法利用门控循环单元学习用户行为。首先，利用XGBoost进行特征提取，克服了传统方法中数据模型复杂的缺陷，使数据模型在保持原始属性的基础上大大降低了复杂度；其次，利用改进后的Dropout网络对数据进行处理，使得召回率提高了1.32%；最后，基于Learning to Rank与Pairwise方法训练用户会话数据，尽可能为用户提供一个与查询内容相关性较强的正向排序推荐清单。实验在Trivago RecSys Challenge 2019数据集上进行。结果表明，所提出的推荐算法在召回率和平均倒数排名上均有提高，而且可以应用于大规模数据推荐。

Abstract: Based on users’ session logs, a TF-Ranking recommendation method that integrated XGBoost and gated recurrent unit was proposed. This method used gated recurrent unit to learn user session data. Firstly, XGBoost was used to extract features, and experimental results showed that XGBoost could overcome the defects of traditional data model and greatly reduced the complexity of data models while maintaining their original attributes. Secondly, an improved Dropout network was used to process data, leading to a recall rate improved by 1.32%. Finally, by training data based on Learning to Rank method in combination with pairwise method, a positive sort recommendation list with strong relevance to query contents was provided for users. Experiments were conducted on the data set of Trivago Recsys Challenge 2019. The results show that the proposed algorithm could improve the recall rate and the average reciprocal ranking, and could be applied to large-scale data recommendation.

Key words: recommender system, TF-Ranking recommendation, gated recurrent unit, boosting tree model, Dropout network, dwell time

中图分类号:

TP391

贾丹, 孙静宇. 基于用户会话的TF-Ranking推荐方法[J]. 应用科学学报, 2021, 39(3): 495-494.

JIA Dan, SUN Jingyu. TF-Ranking Recommendation Method Based on User Session[J]. Journal of Applied Sciences, 2021, 39(3): 495-494.

参考文献

[1] 王卫平, 刘颖. 基于用户行为序列的推荐算法[J]. 计算机系统应用, 2006: 35-38. Wang W P, Liu Y. Recommendation algorithm based on customer behavior locus[J]. Computer Systems and Applications, 2006: 35-38. (in Chinese)
[2] 张永锋. 推荐系统调研报告及综述[EB/OL]. (2016-03-25)[2020-05-08]. https://wenku.baidu.com/view/0cd314cd0722192e4436f6aa.html.2016-03-25.
[3] 陈琛. Netflix大赛电影推荐算法[EB/OL]. https://zhuanlan.zhihu.com/p/74489418.2019-07-23.
[4] Covington P, Adams J, Sargin E. Deep neural networks for YouTube recommendations[C]//Proceedings of the 10th ACM Conference on Recommender Systems. New York: ACM, 2016: 191-198.
[5] Hidasi B, Karatzoglou A, Baltrunas L, et al. Session-based recommendations with recurrent neural networks[C]//ICLR: the International Conference on Learning Representations. US: Computer Science, Mathematics, CoRR, abs/1511.06939. 2016: 319-325.
[6] Yong K T, Xu X X, Liu Y. Improved recurrent neural networks for session-based recommendations[C]//Proceedings of the 1st Workshop on Deep Learning for Recommender Systems. United States: ACM, 2016: 17-22.
[7] Pasumarthi R K, Wang X H, Bruch S. TF-Ranking: scalable tensorflow library for learning-to-rank[EB/OL]. (2018-12-05)[2020-05-08]. http://ai.googleblog.com/2018/12/tfranking-scalable-tensorflow-library.html.
[8] 蔡海尼, 牛冰慧, 文俊浩, 等. 基于时序模型和矩阵分解的推荐系统算法[J]. 计算机应用研究, 2018, 35(6): 1624-1627. Cai H N, Nui B H, Wen J H, et al. Recommender algorithm based on time series model and matrix factorization[J]. Application Research of Computers, 2018, 35(6): 1624-1627. (in Chinese)
[9] Yu Z P, Lian J X, Mahmoody A, et al. Adaptive user modeling with long and short-term preferences for personalized recommendation[C]//IJCAI-19: the Twenty-Eighth International Joint Conference on Artificial Intelligence. Burlington: Morgan Kaufmann, 2019: 4213-4219.
[10] 陈俊航, 徐小平, 杨恒泓. 基于Seq2seq模型的推荐应用研究[J]. 计算机科学, 2019, 46(6): 493-496. Chen J H, Xu X P, Yang H H. Research on recommendation application based on Seq2seq model[J]. Computer Science, 2019, 46(6): 493-496. (in Chinese)
[11] Dallmann A, Grimm A, Plitz C, et al. Improving session recommendation with recurrent neural networks by exploiting dwell time[DB/OL].[2020-05-08]. https://arxiv.org/abs/1706.10231.
[12] 应毅, 刘亚军, 陈诚. 基于云计算技术的个性化推荐系统[J]. 计算机工程与应用, 2015, 51(13): 111-117. Ying Y, Liu Y J, Chen C. Personalization recommender system based on cloud-computing technology[J]. Computer Engineering and Applications, 2015, 51(13): 111-117. (in Chinese)
[13] Chen T Q, Guestrin C. XGBoost: a scalable tree boosting system[C]//KDD’16: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York: ACM, 2016: 785-794.
[14] Tok A. The montonic, darting, best-first boosted tree[EB/OL]. (2019-07-08)[2020-05-08]. http://link.medium.com/g2GtCYafJ4.
[15] 张祖平, 沈晓阳. 基于深度学习的用户行为推荐方法研[J]. 计算机工程与应用, 2019, 55(4): 142-147, 158. Zhang Z P, Shen X Y. Research on user behavior recommendation method based on deep learning[J]. Computer Engineering and Applications, 2019, 55(4): 142-147, 158. (in Chinese)
[16] Lee H, Ahn Y, Lee H, et al. Quote recommendation in dialogue using deep neural network[C]//Proceedings of the 39th International ACM SIGIR Conference on Research and Development in Information Retrieval. United States: ACM, 2016: 957-960.
[17] Tan Y K, Xu X X, Liu Y. Improved recurrent neural networks for session-based recommendations[C]//DLRS 2016: Workshop on Deep Learning for Recommender Systems. New York: Association for Computing Machinery, 2016: 17-22.
[18] Messica A, Rokach L, Friedman M. Session-based recommendations using item embedding[C]//IUI ’17: Proceedings of the 22nd International Conference on Intelligent User Interfaces. New York: ACM, 2017: 629-633.
[19] Hidasi B, Karatzoglou A, Baltrunas L, et al. Recurrent neural networks with top-k gains for session-based recommendations[C]//Proceedings of the 27th ACM International Conference on Information and Knowledge Management. New York: ACM, 2018: 843-852.
[20] Jalili M. A survey of collaborative filtering recommender algorithms and their evaluation metrics[J]. International Journal of System Modeling and Simulation, 2017: 2(2): 14-17.

基于用户会话的TF-Ranking推荐方法

TF-Ranking Recommendation Method Based on User Session

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 1

编辑推荐

Metrics

本文评价