融合用户兴趣漂移的 Top-N 推荐算法

中国民航大学学报 ›› 2021, Vol. 39 ›› Issue (3): 56-61.

融合用户兴趣漂移的 Top-N 推荐算法

刘浩翰，马晓璐，贺怀清

（中国民航大学计算机科学与技术学院，天津 300300）

收稿日期:2020-06-22 修回日期:2020-06-22 接受日期:2020-05-15 出版日期:2021-06-25 发布日期:2021-11-28
作者简介:刘浩翰（1966—），男，黑龙江富锦人，副教授，硕士，研究方向为民航信息智能处理

Top-N recommendation algorithm based on user’s interest drift

LIU Haohan, MA Xiaolu, HE Huaiqing

（College of Computer Science and Technology, CAUC, Tianjin 300300, China）

Received:2020-06-22 Revised:2020-06-22 Accepted:2020-05-15 Online:2021-06-25 Published:2021-11-28

摘要/Abstract

摘要： 传统推荐算法多基于用户兴趣的静态属性获得用户偏好，忽略了用户兴趣漂移问题，为此，提出了解决该问题的融合用户兴趣漂移的 Top-N 推荐算法。利用长短期记忆网络（LSTM，long short term memory）处理时序数据的优势表示用户短期兴趣漂移规律，用矩阵分解得到的固定向量表示用户的长期兴趣，将注意力机制纳入 LSTM 隐藏状态的表示中来获取用户长短期兴趣关联。实验结果表明，所提算法与当前流行算法相比，在 Top-N 项目推荐中具有更优性能。

Abstract: Traditional recommendation algorithms are mostly based on the static attributes of users without considering the drift of user’s interest. To this end, Top-N recommendation algorithm integrating user’s interest drift detection is proposed. The new algorithm makes full use of long short-term memory(LSTM) in processing time series data to represent short -term interest shift of users, compromises a fixed vector obtained by matrix factorization to represent long-term interest of users, and incorporates the attention mechanism into the representation of hidden state of the long short-term memory network to obtain the effect of the user’s long-term interest on the shortterm interest. Compared with current popular algorithms, performance of the proposed algorithm is superior in Top-N item recommendation.

Key words: recommendation algorithm, long short-term memory(LSTM), interest drift, matrix factorization, attention mech鄄 anism, time dynamics

中图分类号:

TP301

刘浩翰, 马晓璐, 贺怀清. 融合用户兴趣漂移的 Top-N 推荐算法 [J]. 中国民航大学学报, 2021, 39(3): 56-61.

LIU Haohan, MA Xiaolu, HE Huaiqing. Top-N recommendation algorithm based on user’s interest drift [J]. Journal of Civil Aviation University of China, 2021, 39(3): 56-61.

参考文献 11

[1]	KOREN Y, BELL R, VOLINSKY C. Matrix factorization techniques forrecommender systems[J]. Computer, 2009, 42(8): 30-37.
[2]	KOREN Y. Collaborative filtering with temporal dynamics[J]. Communications of the ACM, 2010, 53(4): 89-97.
[3]	STEFFEN R, CHRISTOPH F, ZENO G, et al. BPR: Bayesian personalized ranking from implicit feedback[C]//The 25th Conference on Uncertainty in Artificial Intelligence, January 18-21, 2009, Montreal, Canada,2009: 452-461.
[4]	WU C Y, AMR A, ALEX B, et al. Recurrent recommender networks[C]//The 10th ACM International Conference on Web Search and Data Mining, February 6 -10, 2017, Cambridge, UK. New York, USA: MAC,2017: 495-503.
[5]	HIDASI B, KARATZOGLOU A, BALTRUNAS L, et al. Session-basedrecommendations with recurrent neural networks[C]//International Conference on Learning Representations, May 2-4, 2016, San Juan, PuertoRico.
[6]	HIDASI B, KARATZOGLOU A. Recurrent neural networks with top-kgains for session-based recommendations[C]//The 27th ACM International Conference on Information and Knowledge Management, October22-26, 2018, Torino, Italy. New York, USA: ACM, 2018: 843-852.
[7]	ZHOU C, BAI J Z, SONG J S, et al. ATRank: an attention-based userbehavior modeling frameworkfor recommendation[C]//The Associationfor the Advance of Artificial Intelligence, February 2-7, 2018, New Orleans, America.
[8]	WANG T, CHO K, WEN M S. Attention-based mixture density recurrent networks for history-based recommendation[C]//The 1st InternationalWorkshop on Deep Learning Practice for High -DimensionalSparse Data, Anchorage Alaska. New York, USA: ACM, 2019: 1-9.
[9]	ASHISH V, NOAM S, NIKI P, et al. Attention is all you need[C]//Advances in Neural Information Processing Systems, Long Beach California, America, December 4-10, 2017: 5998-6008.
[10]	杨苏雁. 基于矩阵分解及深度神经网络的推荐排序学习方法研究[D]. 北京: 北京交通大学, 2018.
[11]	王俊淑, 张国明, 胡斌. 基于深度学习的推荐算法研究综述[J]. 南京师范大学学报(工程技术版), 2018, 18(4): 33-43.