基于门控循环单元与主动学习的协同过滤推荐算法

doi:10.6040/j.issn.1672-3961.0.2019.411

摘要/Abstract

摘要：

针对传统协同过滤推荐算法存在无法反映用户短时兴趣的问题,提出一种基于门控循环单元(gated recurrent unit, GRU)神经网络与主动学习的协同过滤推荐算法。在采用GRU神经网络的基础上,将数据进行时序化处理,反映用户兴趣变化,并利用主动学习动态采样数据中的高质量的数据进行GRU神经网络的训练,使模型快速建立。在MovieLens1M数据集上的试验结果表明:加入主动学习的GRU模型的推荐算法比基于用户的协同过滤推荐算法(user-based collaborative filtering, UCF)、基于马尔科夫模型的协同过滤推荐算法(markov chain, MC)、基于隐语义模型的协同过滤推荐算法(latent factor model, LFM)算法有更高的短时预测率、召回率、项目覆盖率以及用户覆盖数,能够有效预测用户短时兴趣,提升精度,发掘长尾物品,且与原始GRU模型相比能够以更少的迭代次数达到相同效果。

关键词: 协同过滤, 门控循环单元, 主动学习, 深度学习, 时序化数据

Abstract:

The traditional collaborative filtering recommendation algorithm failed to reflect short-term user interest. In order to reflect the short-term interests of users better, a collaborative filtering recommendation algorithm based on Gated Recurrent Unit (GRU) neural network with active learning was proposed. Based on the GRU neural network, the algorithm processed the data into time-series data to reflect the change of the user's interest and used active learning to sample high-quality data dynamically for accelerating the training of GRU neural network. The result on MovieLens1M dataset showed that the GRU model with active learning could obtain higher short-term prediction success rate, recall rate, item coverage, and user coverage compared with the user-based collaborative filtering method (UCF), the markovian chain model-based collaborative filtering method (MC) and the matrix factory-based collaborative filtering method (LFM), so it could effectively predict the short-term interest of users, improve the accuracy, discover the long-tail items. Meanwhile, it could achieve the same effect with fewer iterations compared with the original GRU model.

Key words: collaborative filtering, gated recurrent unit, active learning, deep learning, time-series data

中图分类号:

TP311

陈德蕾, 王成, 陈建伟, 吴以茵. 基于门控循环单元与主动学习的协同过滤推荐算法[J]. 山东大学学报 (工学版), 2020, 50(1): 21-27.

Delei CHEN, Cheng WANG, Jianwei CHEN, Yiyin WU. GRU-based collaborative filtering recommendation algorithm with active learning[J]. Journal of Shandong University(Engineering Science), 2020, 50(1): 21-27.

图/表 8

图1

表1

表2

表3

图2

图3

图4

图5

参考文献 27

1	冷亚军, 陆青, 梁昌勇. 协同过滤推荐技术综述[J]. 模式识别与人工智能, 2014, 27 (8): 720- 734. doi: 10.3969/j.issn.1003-6059.2014.08.007
	LENG Yajun , LU Qing , LIANG Changyong . Survey of recommendation based on collaborative filtering[J]. Pattern Recognition and Artificial Intelligence, 2014, 27 (8): 720- 734. doi: 10.3969/j.issn.1003-6059.2014.08.007
2	翁小兰, 王志坚. 协同过滤推荐算法研究进展[J]. 计算机工程与应用, 2018, 54 (1): 25- 31.
	WENG Xiaolan , WANG Zhijian . Research progress of collaborative filtering recommendation algorithm[J]. Computer Engineering and Applications, 2018, 54 (1): 25- 31.
3	KOREN Y , BELL R . Recommender systems handbook[M]. New York, USA: Springer, 2015: 77- 118.
4	THORAT P B , GOUDAR R M , BARVE S . Survey on collaborative filtering, content-based filtering and hybrid recommendation system[J]. International Journal of Computer Applications, 2015, 110 (4): 31- 36. doi: 10.5120/19308-0760
5	ELAHI M , RICCI F , RUBENS N . A survey of active learning in collaborative filtering recommender systems[J]. Computer Science Review, 2016, 20, 29- 50. doi: 10.1016/j.cosrev.2016.05.002
6	季芸, 胡雪蕾. 基于Baseline SVD主动学习算法的推荐系统[J]. 现代电子技术, 2015, 38 (12): 8- 11.
	JI Yun , HU Xuelei . Recommender system based on Baseline SVD active learning algorithm[J]. Modern Electronics Technique, 2015, 38 (12): 8- 11.
7	余天豪.基于社会网络的主动信息推送算法研究[D].杭州:杭州师范大学, 2012.
	YU Tianhao. Research on recommendation algorithm based on social network[D]. Hangzhou: Hangzhou Normal University, 2012.
8	GUO G, ZHANG J, YORKE-SMITH N. Trust SVD: collaborative filtering with both the explicit and implicit influence of user trust and of item ratings[C]//Proceedings of the Twenty-Ninth AAAI Conference on Artificial Intelligence. Texas, USA: AAAI Press, 2015: 123-129.
9	JOHNSON J, NG Y K. Enhancing long tail item recommendations using tripartite graphs and Markov process[C]//Proceedings of the International Conference on Web Intelligence. California, USA: ACM, 2017: 761-768.
10	ALSHAMMARI G, JORRO-ARAGONESES J L, KAPETANAKIS S, et al. A hybrid CBR approach for the long tail problem in recommender systems[C]// International Conference on Case-Based Reasoning. Trondheim, Norway: Springer, 2017: 35-45.
11	HE X, LIAO L, ZHANG H, et al. Neural collaborative filtering[C]// Proceedings of the 26th International Conference on World Wide Web. Perth, Australia: ACM, 2017: 173-182.
12	XUE H J, DAI X Y, ZHANG J, et al. Deep matrix factorization models for recommender systems[C]//International Joint Conference on Artificial Intelligence. Melbourne, Australia: AAAI Press, 2017: 3203-3209.
13	CHEN T, SUN Y, SHI Y, et al. On sampling strategies for neural network-based collaborative filtering[C]//Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Halifax, Canada: ACM, 2017: 767-776.
14	BILLSUS D, PAZZANI M J. Learning collaborative information filters[C]// Proceedings of the Fifteenth International Conference on Machine Learning. California, USA: Morgan Kaufmann Publishers Inc., 1998: 46-54.
15	FUNK S. Netflix update: try this at home[EB/OL]. (2006-12-11) [2019-04-12]. http://sifter.org/~simon/journal/20061211.html.
16	KOREN Y . Factor in the neighbors: scalable and accurate collaborative filtering[J]. Acm Transactions on Knowledge Discovery from Data, 2010, 4 (1): 1- 24.
17	DEVOOGHT R, BERSINI H. Long and short-term recommendations with recurrent neural networks[C]//Proceedings of the 25th Conference on User Modeling, Adaptation and Personalization. Bratislava, Slovakia: ACM, 2017: 13-21.
18	LIU J , WU C , WANG J . Gated recurrent units based neural network for time heterogeneous feedback recommendation[J]. Information Sciences, 2018, 423, 50- 65. doi: 10.1016/j.ins.2017.09.048
19	HOCHREITER S , SURHONE J . Long short-term memory[J]. Neural Computation, 1997, 9 (8): 1735- 1780. doi: 10.1162/neco.1997.9.8.1735
20	CHO K, VAN MERRIENBOER B, GULCEHRE C, et al. Learning phrase representations using RNN encoder-decoder for statistical machine translation[C]// Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing. Doha, Qatar: MIT Press, 2014: 1724-1734.
21	HIDASI B, KARATZOGLOU A. Recurrent neural networks with top-k gains for session-based recommendations[C]//Proceedings of the 27th ACM International Conference on Information and Knowledge Management. Torino, Italy:ACM, 2018: 843-852.
22	CHAKRABORTY S , BALASUBRAMANIAN V , SUN Q . Active batch selection via convex relaxations with guaranteed solution bounds[J]. IEEE Trans Pattern Anal Mach Intell, 2015, 37 (10): 1945- 1958. doi: 10.1109/TPAMI.2015.2389848
23	KONYUSHKOVA K, SZNITMAN R, FUA P. Learning active learning from data[C]//Advances in Neural Information Processing Systems. California, USA: ACM, 2017: 4225-4235.
24	HUANG S J, ZHAO J W, LIU Z Y. Cost-effective training of deep cnns with active model adaptation[C]//Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. London, UK:ACM, 2018: 1580-1588.
25	KARIMI R, FREUDENTHALER C, NANOPOULOS A, et al. Non-myopic active learning for recommender systems based on Matrix Factorization[C]//IEEE International Conference on Information Reuse and Integration. Las Vegas, USA: IEEE, 2011:299-303.
26	ZHOU D, WANG B, RAHIMI S M, et al. A study of recommending locations on location-based social network by collaborative filtering[C]//Canadian Conference on Artificial Intelligence. Toronto, Canada:Springer, 2012: 255-266.
27	RENDLE S, FREUDENTHALER C, SCHMIDT-THIEME L. Factorizing personalized markov chains for next-basket recommendation[C]//Proceedings of the 19th international conference on World wide web. North Carolina, USA:ACM, 2010: 811-820.

多维度评价

Viewed

Full text

Abstract

Cited

Shared

Discussed

指标	UCF	MC	LFM	GRU+Active
使用数据	原始评分矩阵	时序化评分数据	原始评分矩阵	时序化评分数据
反映短时用户兴趣	差	一般	差	好
短时预测精度	低	一般	低	高
发掘长尾物品能力	差	一般	差	好

方法	参数
GRU+随机采样	l=0.05, k=50, adam
GRU+Active	a=100, l=0.05, k=50, adam

方法	Sps/%	Recall/%	Item-coverage/个	User-coverage/%
UCF	11.62	5.77	301	77.99
LFM	11.28	5.65	399	80.80
MC	19.54	5.27	558	78.70
GRU	32.16	8.85	659	87.98
GRU+Active	32.51	8.86	662	88.81

[1]	李常刚,李宝亮,曹永吉,王佳颖. 人工智能在电力系统潮流计算中的应用综述及展望[J]. 山东大学学报 (工学版), 2025, 55(5): 1-17.
[2]	周群颖,隋家成,张继,王洪元. 基于自监督卷积和无参数注意力机制的工业品表面缺陷检测[J]. 山东大学学报 (工学版), 2025, 55(4): 40-47.
[3]	薛冰冰,王勇,杨维浩,王川,于迪,王旭. 基于ETC收费数据的高速公路交通流数据修复及实时预测[J]. 山东大学学报 (工学版), 2025, 55(3): 58-71.
[4]	董明书,陈俐企,马川义,张珠皓,孙仁娟,管延华,庄培芝. 沥青路面内部裂缝雷达图像智能判识算法研究[J]. 山东大学学报 (工学版), 2025, 55(3): 72-79.
[5]	常新功,苏敏惠,周志刚. 基于进化集成的图神经网络解释方法[J]. 山东大学学报 (工学版), 2024, 54(4): 1-12.
[6]	索大翔,李波. 基于Gromov-Wasserstein最优传输的输电线路小目标检测方法[J]. 山东大学学报 (工学版), 2024, 54(3): 22-29.
[7]	宋辉,张轶哲,张功萱,孟元. 基于类权重和最小化预测熵的测试时集成方法[J]. 山东大学学报 (工学版), 2024, 54(3): 36-43.
[8]	刘新,刘冬兰,付婷,王勇,常英贤,姚洪磊,罗昕,王睿,张昊. 基于联邦学习的时间序列预测算法[J]. 山东大学学报 (工学版), 2024, 54(3): 55-63.
[9]	聂秀山,巩蕊,董飞,郭杰,马玉玲. 短视频场景分类方法综述[J]. 山东大学学报 (工学版), 2024, 54(3): 1-11.
[10]	徐金华,罗义凯,李昱燃,李岩. 基于时频分解与深度学习的轨道客流预测[J]. 山东大学学报 (工学版), 2024, 54(2): 60-68.
[11]	李璐,张志军,范钰敏,王星,袁卫华. 面向冷启动用户的元学习与图转移学习序列推荐[J]. 山东大学学报 (工学版), 2024, 54(2): 69-79.
[12]	高泽文,王建,魏本征. 基于混合偏移轴向自注意力机制的脑胶质瘤分割算法[J]. 山东大学学报 (工学版), 2024, 54(2): 80-89.
[13]	陈成,董永权,贾瑞,刘源. 基于交互序列特征相关性的可解释知识追踪[J]. 山东大学学报 (工学版), 2024, 54(1): 100-108.
[14]	李家春,李博文,常建波. 一种高效且轻量的RGB单帧人脸反欺诈模型[J]. 山东大学学报 (工学版), 2023, 53(6): 1-7.
[15]	王旭晴,魏伟波,杨光宇,宋金涛,吕婷,潘振宽. 基于算法展开的图像盲去模糊深度学习网络[J]. 山东大学学报 (工学版), 2023, 53(6): 35-46.