数据稀疏环境下基于用户主题偏好的协同过滤算法

运筹与管理 ›› 2014, Vol. 23 ›› Issue (2): 145-152.

数据稀疏环境下基于用户主题偏好的协同过滤算法

张尧, 冯玉强

哈尔滨工业大学信息管理与信息系统研究所,黑龙江哈尔滨 150001

收稿日期:2012-06-12 出版日期:2014-02-25
作者简介:张尧(1981-),男,黑龙江哈尔滨人,博士研究生,研究方向:电子商务、电子推荐、网络信任;冯玉强(1961-),女,黑龙江哈尔滨人,教授,博士生导师,研究方向:商务谈判、数据挖掘、企业信息化、客户满意度。
基金资助:
国家自然科学基金资助项目(71172157);国家自然科学基金海外合作基金(71028003)

Topic Preference Based Method for Collaborative Filtering Algorithm in Sparse Datasets

ZHANG Yao, FENG Yu-qiang

School of Management, Harbin Institute of Technology, Harbin 150001, China

Received:2012-06-12 Online:2014-02-25

摘要/Abstract

摘要： 在B2C电子商务中,user-based协同过滤算法是一种重要的推荐方法,但用户共同评价项目数据稀疏影响了user-based协同过滤算法的应用。鉴于此,在考虑用户消费水平的基础上,利用关联规则挖掘形式化描述商品间的替代相似性;利用基于时间的贝叶斯概率描述商品间的关联关系构建商品网络,通过社会网络分析中的成份分析方法对商品网分析,得到面向用户主题偏好的商品间互补性关系,进而利用这两种商品间关系构建用户主题偏好项目集,最后在数据极度稀疏的环境下通过F₁方法和多样性测量方法与传统推荐算法进行对比实验分析,实验结果显示提高了推荐结果的准确性与新颖性。研究用的所有数据均采集于京东商城网站。本文为缓解数据稀疏问题提出了一种新的方法,扩展了整体网分析方法在商品关系分析中的应用,含有理论与实践双重意义。

关键词: B2C, 协同过滤, 稀疏问题, 主题偏好, 商品关系, 社会网络分析

Abstract: User-based collaborative filtering algorithm is an important method for B2C electronic commerce to recommend commodity, but it has been limited to some extent because of the sparsity of common rations between users. In order to resolve these problems,the paper first adopts association rule mining to formalize similarity among competitive goods based on considering consumption level, and then constructs time-based Bayesian goods relation network, based on the network the paper takes advantage of components analysis of whole network to find complementary similarity of the goods and the topic preference of consumer for expanding common rating sets. At last, through comparative experiments based on F₁ method and diversity method, the result shows that the accuracy and diversity have been improved significantly in sparsity environment. The data are collected from the site of JingDong Mall. In conclusion, the model provides a new way of dealing with sparsity problem, enriches examples using whole network approach in goods relationship analysis, and has the significance in theory and practice.

Key words: B2C, collaborative filtering, sparsity problem, topic preference, goods relationship, social network analysis

中图分类号:

TP391

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.

ZHANG Yao, FENG Yu-qiang. Topic Preference Based Method for Collaborative Filtering Algorithm in Sparse Datasets[J]. Operations Research and Management Science, 2014, 23(2): 145-152.

参考文献

[1] 洪兴文,翁洋,朱顺痣,李茂青.垂直电子商务网站的混合型推荐系统[J].系统工程理论与实践,2010,30(5):928-935.
[2] O'sullivan D, Smyth B, Wilson D. Preserving recommender accuracy and diversity in sparse datasets[J]. International Journal on Artificial Intelligence Tools, 2004 ,13(1): 219-235.
[3] Vozalis M G, Margaritis K G. Using SVD and demographic data for the enhancement of generalized collaborative filtering[J]. Information Sciences, 2007, 177(15): 3017-3037.
[4] Goldberg K, Roeder T. Eigentaste: a constant time collaborative filtering algorithm[J]. Information Retrieval 2001, 4(2): 133-151.
[5] Bridge D, Kelleher J. Experiments in sparsity reduction: using clustering in collaborative recommenders[J]. Artificial Intelligence and Cognitive Science, 2002, 2464. 5-34.
[6] 邓爱林,朱阳勇,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.
[7] 李聪,梁昌勇,马丽.基于领域最近邻的协同过滤推荐算法[J].计算机研究与发展,2008,45(9):1532-1538.
[8] 王茜,杨莉云,杨德礼.面向用户偏好的属性值评分分布协同过滤算法[J].系统工程学报,2010,25(4):561-568.
[9] 李永利,吴冲,王崑声.基于图论和信息最大化保留的在线推荐方法[J].系统工程理论与实践,2011,31(9):1718-1725.
[10] Scott J. Social network analysis[M]. A Handbook. London: Sage Publications, 2000.
[11] Yang Y, Liu X. A Re-examination of text categorization methods[C]. // Proceeding of SIGIR'99 Proceedings of the 22nd annual international ACM SIGIR conference on Research and development in information retrieval, New York, NY, USA: ACM Press, 1999. 42-49.
[12] Sarwar B M, Karypis G, Konstan J, Riedl J. Analysis of recommendation algorithms for e-commerce[C]. // Proceeding of the 2nd ACM Conference on Electronic Commerce. New York: ACM Press, 2000. 158-167.