基于用户和项目的双视角协同过滤推荐方法

doi:10.3969/j.issn.0255-8297.2017.03.006

应用科学学报 ›› 2017, Vol. 35 ›› Issue (3): 326-336.doi: 10.3969/j.issn.0255-8297.2017.03.006

基于用户和项目的双视角协同过滤推荐方法

程树林^1,2, 张博锋¹, 邹国兵¹

1. 上海大学计算机工程与科学学院, 上海 200444;
2. 安庆师范大学计算机与信息学院, 安徽安庆 246133

收稿日期:2016-02-28 修回日期:2016-10-03 出版日期:2017-05-30 发布日期:2017-05-30
作者简介:程树林，博士生，副教授，研究方向：个性化推荐，E-mail：chengshulin@shu.edu.cn;张博锋，研究员，博导，研究方向：智能信息处理、个性化推荐，E-mail：bfzhang@shu.edu.cn
基金资助:
国家自然科学基金（No.61303096）；上海市自然科学基金（No.13ZR1454600）资助

Collaborative Filtering Recommendation Based on Double-Perspective of Users and Items

CHENG Shu-lin^1,2, ZHANG Bo-feng¹, ZOU Guo-bing¹

1. School of Computer and Science Engineering, Shanghai University, Shanghai 200444, China;
2. Institute of Computer and Information, Anqing Normal University, Anqing 246133, Anhui Province, China

Received:2016-02-28 Revised:2016-10-03 Online:2017-05-30 Published:2017-05-30

摘要/Abstract

摘要：

传统的协同过滤推荐方法存在单视角信息利用不足、预测精度不高、对数据稀疏性敏感等问题，为此提出同时考虑相似用户和相似项目的双视角协同过滤推荐方法。根据辩证的思想，利用项目内部因子和外部因子生成项目融合相似度，有效度量了项目相似性和用户相似性，并解决了双视角协同过滤推荐方法对数据稀疏性敏感的问题。在标准数据集上多次进行的实验表明，基于用户和项目的双视角协同过滤推荐方法优于多个典型的协同过滤推荐方法。

关键词: 双视角, 协同过滤推荐, 融合相似度

Abstract:

Traditional collaborative fltering (CF) recommendation approach has a serious problems such as insufcient usage of single perspective information, unsatisfactory accuracy and sensitivity to data sparsity. To solve these problems, a CF recommendation method based on double-perspective of users and items is proposed by considering information of similar users and similar items. According to the dialectic principle, fusion similarity of items is given by combination of inner-factors and outer-factors of the item. This way, the item similarity and user similarity can be effectively measured. The measurement is robust against data sparsity in the approach of CF recommendation based on double-perspective of user and item. Several experiments are carried on benchmark datasets. The results show that the proposed CF recommendation method based on double-perspective of users and items outperforms several other typical CF approaches.

Key words: collaborative fltering recommendation, double-perspective, fusion similarity of item

中图分类号:

TP181

程树林, 张博锋, 邹国兵. 基于用户和项目的双视角协同过滤推荐方法[J]. 应用科学学报, 2017, 35(3): 326-336.

CHENG Shu-lin, ZHANG Bo-feng, ZOU Guo-bing. Collaborative Filtering Recommendation Based on Double-Perspective of Users and Items[J]. Journal of Applied Sciences, 2017, 35(3): 326-336.

参考文献

[1] Ricci F, Rokach L, Shapira B. Introduction to recommender systems handbook[M]. New York:Springer, 2011.
[2] Shi Y, Larson M, Hanjalic A. Collaborative fltering beyond the user-item matrix:a survey of the state of the art and future challenges[J]. ACM Computing Surveys, 2014, 47(1):1-45.
[3] Park D H, Kim H K, Choi I Y, Kim, J K. A literature review and classifcation of recommender systems research[J]. Expert Systems with Applications, 2012, 39(11):10059-10072.
[4] Linden G, Smith B, York J. Amazon.com recommendations:Item-to-item collaborative fltering[J]. IEEE on Internet Computing, 2003, 7(1):76-80
[5] 邓爱林,朱扬勇,施伯乐. 基于项目评分预测的协同过滤推荐算法[J]. 软件学报,2003, 14(9):1621-1628. Deng A L, Zhu Y Y, Shi B L. A collaborative fltering recommendation algorithm based on item rating prediction[J]. Journal of Software, 2003, 14(9):1621-1628(in Chinese).
[6] Ma H, King I, Lü M R. Effective missing data prediction for collaborative fltering[C]//Proceedings of the 30th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval. ACM, 2007:39-46.
[7] Lu Z, Dou Z, Lian J, Xie X, Yang Q. Content-based collaborative fltering for news topic recommendation[C]//29th AAAI Conference on Artifcial Intelligence, 2015:217-233.
[8] Song R P, Wang B, Huang G M, Liu Q D, Hu R J, Zhang R S. A hybrid recommender algorithm based on an improved similarity method[J]. Applied Mechanics and Materials, 2014, 475:978-982.
[9] Moin A, Ignat C L. Hybrid weighting schemes for collaborative fltering[D]. Paris:INRIA Nancy, 2014.
[10] Goldberg D, Nichols D, Oki B M, Terry D. Using collaborative fltering to weave an information tapestry[J]. Communications of the ACM, 1992, 35(12):61-70.
[11] Breese J S, Heckerman D, Kadie C. Empirical analysis of predictive algorithms for collaborative fltering[C]//Proceedings of the Fourteenth Conference on Uncertainty in Artifcial Intelligence. Morgan Kaufmann Publishers Inc., 1998:43-52.
[12] Deshpande M, Karypis G. Item-based top-n recommendation algorithms[J]. ACM Transactions on Information Systems, 2004, 22(1):143-177.
[13] Rennie J D M, Srebro N. Fast maximum margin matrix factorization for collaborative prediction[C]//Proceedings of the 22nd International Conference on Machine Learning, 2005:713-719.
[14] Choi K, Suh Y. A new similarity function for selecting neighbors for each target item in collaborative fltering[J]. Knowledge-Based Systems, 2013, 37:146-153.
[15] Forsati R, Mahdavi M, Shamsfard M, Sarwat M. Matrix factorization with explicit trust and distrust relationships[J]. ArXiv:1408. 0325Ⅵ[cs.SI], 2014.
[16] Ghazanfar M A, Prugel A. The advantage of careful imputation sources in sparse dataenvironment of recommender systems:Generating improved svd-based recommendations[J]. Informatica, 2013, 37(1):61-92.
[17] Anand D, Bharadwaj K K. Pruning trust-distrust network via reliability and risk estimates for quality recommendations[J]. Social Network Analysis and Mining, 2013, 3(1):65-84.
[18] Agarwal V, Bharadwaj K K. A collaborative fltering framework for friends recommendation in social networks based on interaction intensity and adaptive user similarity[J]. Social Network Analysis and Mining, 2013, 3(3):359-379.
[19] Li W, Ye Z, Xin M, Jin Q. Social recommendation based on trust and influence in SNS environments[J]. Multimedia Tools and Applications, 2015:1-18.
[20] Odi?A, Tkal?i?, Tasi? J F, Košir A. Predicting and detecting the relevant contextual information in a movie-recommender system[J]. Interacting with Computers, 2013, 25(1):74-90.
[21] Shani G, Gunawardana A. Evaluating recommendation systems[M]. New York:Springer, 2011:257-297.
[22] Yang X, Guo Y, Liu Y, Steck H. A survey of collaborative fltering based social recommender systems[J]. Computer Communications, 2014, 41:1-10.
[23] Zenebe A, Zhou L, Norcio A F. User preferences discovery using fuzzy models[J]. Fuzzy Sets and Systems, 2010, 161(23):3044-3063.
[24] Wang J, De Vries A P, Reinders M J T. Unifying user-based and item-based collaborative fltering approaches by similarity fusion[C]//Proceedings of the 29th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval. ACM, 2006:501-508.

基于用户和项目的双视角协同过滤推荐方法

Collaborative Filtering Recommendation Based on Double-Perspective of Users and Items

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

本文评价

[1]	代永强, 王联国, 胡斌. 甘肃省地区农业旱灾脆弱性评价[J]. 应用科学学报, 2018, 36(3): 515-523.
[2]	翟旭平, 孟田, 王涛. 采用对数预处理的SVM频谱感知方法[J]. 应用科学学报, 2017, 35(6): 726-734.
[3]	王向阳1,2, 万旺根1,2. 采用多任务稳健主成分分析的运动目标分割[J]. 应用科学学报, 2014, 32(5): 473-480.
[4]	叶飞1;2，衣娜1，王翼飞1. 三阶隐马氏模型算法及其与一阶隐马氏模型的关系[J]. 应用科学学报, 2011, 29(5): 500-507.
[5]	王建国，汪权，张鸣祥. 结构地震响应线性二次型最优迭代学习控制[J]. 应用科学学报, 2010, 28(2): 197-202.