基于聚类矩阵近似的协同过滤推荐研究

doi:10.12005/orms.2020.0104

运筹与管理 ›› 2020, Vol. 29 ›› Issue (4): 171-178.DOI: 10.12005/orms.2020.0104

基于聚类矩阵近似的协同过滤推荐研究

张文^1,3, 崔杨波², 李健¹, 陈进东⁴

1.北京工业大学经济管理学院,北京 100124;
2.北京化工大学经济管理学院,北京100029;
3.西安文理学院信息工程学院,陕西西安 710065;
4.北京信息科技大学经济管理学院,北京 100101

收稿日期:2018-08-20 出版日期:2020-04-25
作者简介:张文(1981-),男,湖北人,教授,博士,研究方向:电子商务与推荐系统;崔杨波(1993-),女,河南人,硕士,研究方向:电子商务;李健(1976-),男,山东人,教授,博士,研究方向:智慧供应链;陈进东(1983-),男,重庆人,副教授,博士,研究方向:数据挖掘。
基金资助:
国家自然科学基金资助项目(71932002,61379046,71601023,61432001);西安市科技计划创新基金项目(2016CXWL21)

F-cluMA: a Collaborative Filtering Model Based on Clustered Matrix Approximation

ZHANG Wen^1,3, CUI Yang-bo², LI Jian¹, CHEN Jin-dong⁴

1. School of Economics and Management, Beijing University of Technology, Beijing 100124, China;
2. School of Economics and Management, Beijing University of Chemical Technology, Beijing 100029, China;
3. School of Information Engineering, Xi'an University, Xi'an 710065, China;
4. School of Economics and Management, Beijing Information Science and Technology University, Beijing 100101, China

Received:2018-08-20 Online:2020-04-25

摘要/Abstract

摘要： 由于推荐系统中存在巨量的用户和商品,现有的协同过滤方法很难处理用户-商品推荐中的数据稀疏性和计算可扩展性问题。本文提出了一种基于聚类矩阵近似的协同过滤推荐方法CF-cluMA。一方面,CF-cluMA方法通过对用户和商品进行分别聚类,并利用聚类后的用户-商品分块评分矩阵来刻画用户对于商品兴趣的局部性特点,以降低用户-商品评分矩阵的全局稀疏性。另一方面,CF-cluMA方法通过对局部稠密分块矩阵实施奇异值分解,并利用施密特变换近似全局用户-商品评分矩阵来预测用户对未知商品评分,以降低协同过滤算法的复杂性。在EachMovie电影评分真实数据集上的实验表明,相比于已有的基于矩阵近似的协同过滤推荐方法,本文所提出的CF-cluMA方法能够有效提升推荐系统的准确性并降低推荐系统的计算复杂性。本文的研究对于电子商务推荐系统具有重要的管理启示。

关键词: 谱聚类, 矩阵近似, 协同过滤, 数据稀疏性, 可扩展性

Abstract: It is extremely hard for the existing collaborative filtering algorithms to handle the problems of data sparsity and computation scalability in a recommendation system. This paper proposes a novel approach called CF-cluMA (Collaborative Filtering based on clustered Matrix Approximation) for collaborative filtering with the goal to tackle the problems. The CF-cluMA approach comprises two components. The first component is to partition the global user-item rating matrix into local dense block matrices by co-clustering its rows (users) and columns (items). The second one is to approximate the global user-item rating matrix by singular value decomposition (SVD) on the local dense blocks with Schimidt orthogonalization. The experiments on the EachMovie dataset demonstrate that the proposed CF-cluMA approach outperforms state-of-the-art collaborative filtering methods in terms of recommendation accuracy and computation complexity. This research has great managerial implications for recommendation systems in E-commerce.

Key words: spectral clustering, matrix approximation, collaborative filtering, data sparsity, computation scalability

中图分类号:

TP18

张文, 崔杨波, 李健, 陈进东. 基于聚类矩阵近似的协同过滤推荐研究[J]. 运筹与管理, 2020, 29(4): 171-178.

ZHANG Wen, CUI Yang-bo, LI Jian, CHEN Jin-dong. F-cluMA: a Collaborative Filtering Model Based on Clustered Matrix Approximation[J]. Operations Research and Management Science, 2020, 29(4): 171-178.

参考文献

[1] 崔春生,王梦冉,王国成.一种基于可拓学的电子商务内容推荐算法研究[J].运筹与管理,2018,27(6):75-81.
[2] Koren K, Bell B, Volinsky C H. Matrix factorization techniques for recommender systems[C]//IEEE Computer, 2009, 42(8): 42-49.
[3] Zhang W, Du Y H, Yoshida T, et al. DeepRec: a deep neural network approach to recommendation with item embedding and weighted loss function[J]. Information Sciences, 2019, 470: 121-140.
[4] Breese J S, Heckerman D, Kadie C. Empirical analysis of predictive algorithms for collaborative filtering[C]//Fourteenth Conference on Uncertainty in Artificial Intelligence. Morgan Kaufmann Publishers Inc. 1998.43-52.
[5] Nguyen J, Zhu M. Content-boosted matrix factorization techniques for recommender systems[J]. Statistical Analysis & Data Mining, 2013, 6(4): 286-301.
[6] Gong S J, Ye H W. Joining user clustering and item based collaborative filtering in personalized recommendation services[C]//Proceedings of International Conference on Industrial and Information Systems, 2009.149-151.
[7] Kumar N P, Fan Z. Hybrid User-item based collaborative filtering[J]. Procedia Computer Science, 2015, 60(1): 1453-1461.
[8] Sarwar Badrul M. Recommender systems for large-scale e-commerce: scalable neighborhood formation using clustering[C]//Proceedings of the Fifth International Conference on Computer and Information Technology. Vol. 1. 2002.
[9] Xia S, ZhaoY, Zhang Y, et al. Optimizations for item-based collaborative filtering algorithm[C]//Proceedings of Natural Language Processing and Knowledge Engineering(NLP-KE), 2010. 1-5.
[10] Gong S J. Joining case-based reasoning and Item-based collaborative filtering in recommender systems[C]//Proceedings of the Second International Symposium on Electronic Commerce and Security, 2009. 40-42.
[11] Chen S, Peng Y. Matrix factorization for recommendation with explicit and implicit feedback[J]. Knowledge-Based Systems, 2018.
[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. AAAI Press, 2017. 3203-3209.
[13] Dhillon I S. Co-clustering documents and words using bipartite spectral graph partitioning[C]//ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, 2001. 269-274.
[14] Golub G H, Loan V, Charles f: matrix computations(third edition)[M]. Johns Hopkins University Press, 1996.
[15] Zhang W, Yoshida T, Tang X J, Wang Q. Text clustering using frequent item sets[J]. Knowledge-based Systems, 2010, 23(5): 379-388.
[16] Savas B, Dhillon I S. Clustered matrix approximation[J]. SIAM Journal on Matrix Analysis and Applications, 2016, 37(4): 1531-1555.
[17] Phuong D, Tu M P. Collaborative filtering by multi-task learning[C]// IEEE International Conference on Research, Innovation and Vision for the Future, 2008.227-232.
[18] Vavasis S A. On the complexity of nonnegative matrix factorization[J]. SIAM Journal on Optimization, 2009, 20(3): 1364-1377.
[19] Zhang W, Du Y H, Yang Y, et al. DeRec: a data-driven approach to accurate recommendation with deep learning and weighted loss function[J]. Electronic Commerce Research and Applications, 2018, 31: 12-23.
[20] Verstrepen K, Bhaduriy B, Cule B, et al. Collaborative filtering for binary, positive-only data[C]// In Proceedings of the 23rd ACM SIGKDD Conference, 2017.1-21.
[21] 陈洁,潘郁,张振海,等.基于用户实时行为的Slope One模型与算法[J].运筹与管理,2015,24(1):89-92

基于聚类矩阵近似的协同过滤推荐研究

F-cluMA: a Collaborative Filtering Model Based on Clustered Matrix Approximation

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 5

编辑推荐

Metrics

[1]	吕成戍. 基于多元隐式信任关系挖掘的抗攻击社会化推荐算法研究[J]. 运筹与管理, 2020, 29(1): 69-78.
[2]	陶维成, 党耀国. 基于灰色关联聚类的协同过滤推荐算法[J]. 运筹与管理, 2018, 27(1): 84-88.
[3]	邰蕾蕾, 陶世奇. 一种基于用户识别特征模型的中医药健康养老信息推送算法[J]. 运筹与管理, 2017, 26(12): 183-188.
[4]	苏木亚. 全球主要股市波动率的聚类特征分析[J]. 运筹与管理, 2017, 26(11): 134-144.
[5]	张尧, 冯玉强. 数据稀疏环境下基于用户主题偏好的协同过滤算法[J]. 运筹与管理, 2014, 23(2): 145-152.