基于灰色关联度聚类与标签重叠因子结合的协同过滤推荐方法研究

J4 ›› 2016, Vol. 38 ›› Issue (01): 171-176.

基于灰色关联度聚类与标签重叠因子结合的协同过滤推荐方法研究

赵宏晨1，翟丽丽1,2，张树臣1,2

(1.哈尔滨理工大学管理学院，黑龙江哈尔滨 150040;
2.哈尔滨理工大学高新技术产业发展研究中心，黑龙江哈尔滨 150040)

收稿日期:2014-09-11 修回日期:2014-12-16 出版日期:2016-01-25 发布日期:2016-01-25
基金资助:
国家自然科学基金（71272191，71072085）；黑龙江省自然科学基金（G201301）；黑龙江省高等学校哲学社会科学创新团队建设计划(TD201203)；黑龙江省博士后基金（LBHZ14068）

A collaborative filtering recommendation method
based on clustering of gray association degree
and factors of tag overlap

ZHAO Hongchen1,ZHAI Lili1,2,ZHANG Shuchen1,2

(1.College of Management,Harbin University of Science and Technology,Harbin 150040;
2.Hightech Industrial Development Research Center,Harbin University of Science and Technology,Harbin 150040,China)

Received:2014-09-11 Revised:2014-12-16 Online:2016-01-25 Published:2016-01-25

摘要/Abstract

摘要：

协同过滤算法是目前被广泛运用在推荐系统领域的最成功技术之一，但是面对用户数量的快速增长及相应的评分数据的缺失，推荐系统中的数据稀疏性问题也越来越明显，严重地影响着推荐的质量和效率。针对传统协同过滤算法中的稀疏性问题，采用了基于灰色关联度的方法对用户评分矩阵进行数据标准化处理，得到用户关联度并形成关联度矩阵；然后对关联矩阵中的用户进行关联度聚类，以减少相似性算法的复杂度；之后利用标签重叠因子对传统计算用户相似性的协同过滤算法进行改进，将重叠因子与用户评分以非线性形式进行组合；最后通过实例改进后的算法在推荐精确度上有着较大的提高。

Abstract:

Collaborative filtering algorithms are one of the most successful techniques which is widely used in the field of recommendation system. However, with the rapid growth in the number of users and the lack of corresponding rating data, the problem of data sparsity in recommender systems is becoming more and more obvious, which affects recommendation quality and efficiency seriously. To solve the sparse data problem in traditional collaborative filtering algorithms, we propose a method based on gray correlation degree, combination of tag overlap factors and user ratings, which is used for the standardization of the data processing of the user rating matrix, thus obtaining the user correlation degree and the association degree matrix. Clustering based on gray association degree on the correlation matrix of users can reduce computational complexity. Tag overlap factors are also introduced to improve the accuracy of the similarity between users in traditional collaborative filtering algorithms. Experimental results show that the proposed algorithm can greatly improve the prediction accuracy.

Key words: collaborative filtering;gray correlation degree;factors of tag overlap

赵宏晨1，翟丽丽1,2，张树臣1,2 . 基于灰色关联度聚类与标签重叠因子结合的协同过滤推荐方法研究[J]. J4, 2016, 38(01): 171-176.

ZHAO Hongchen1,ZHAI Lili1,2,ZHANG Shuchen1,2. A collaborative filtering recommendation method
based on clustering of gray association degree
and factors of tag overlap [J]. J4, 2016, 38(01): 171-176.

参考文献［17］

［1］	Resnick P,Varian H R. Recommender systems［J］. Communications of the ACM,1997,40(3):5658.
［2］	Xu Hailing, Wu Xiao, Li Xiaodong, et al. Comparison study of Internet recommendation system［J］. Journal of Software,2009,20(2):350362.(in Chinese)
［3］	David P,Eric H,Steve L,et al. Collaborative filtering by personality diagnosis:A hybrid memory and modelbased approach［C］∥Proc of the 16th Conference on Uncertainty in Artificial Intelligence,2000:473480.
［4］	Liu Fangxian,Song Shunlin. Improved collaborative filtering recommendation algorithm［J］. Computer Engineering and Applications,2011,47(8):7275.(in Chinese)
［5］	Wei Suyun,Ye Ning. Collaborative filtering recommendation algorithm based on item clustering and global similarity［J］. Computer Science,2012,39(12):149152.(in Chinese)
［6］	Li Cong,Liang Changyong. Coldstart eliminating method of collaborative filtering based on nsequence access analytic logic［J］. Systems Engineering Theory and Practice,2012,32(7):15371545.(in Chinese)
［7］	Xiong Zhongyang,Liu Qin. Improved algorithm of collaborative filtering based on item classification［J］. Application Research of Computer,2012,29(2):493496. (in Chinese)
［8］	Wu Hu,Wang Yongji,Wang Zhe,et al. Twophase collaborative filtering algorithm based on coclustering［J］. Journal of Software,2010,21(5):10421054.(in Chinese)
［9］	Tsai F，Hung C．Cluster ensembles in collaborative filtering recommendation［J］.Applied Soft Computing,2012,12 (4):14171425．
［10］	Sun Xiaohua, Chen Hong, Kong Fansheng. Combining singular value decomposition and neighbor based method in collaboration filtering［J］.Application Research of Computers,2006,23(9):206208.(in Chinese)
［11］	Breese J S,Heeckerman D,Kadie C. Empirical analysis of predictive algorithms for collaborative filtering ［R］. Redmond:Microsoft Research,1998.
［12］	Adomavicius G,Tuzhilin A. Toward the next generation recommender systems:A survey of the stateoftheart and possible extensions［J］. IEEE Transactions on Knowledge and Data Engingeering,2005,17(6):734749.
［13］	Sung H. Helping online customers decide through Web personalization ［J］. IEEE Intelligent Systems,2002(9):3443.
［14］	Deng Ailin. The research on key technologies of recommendation system in Ecommerce［D］. Shanghai:Fudan University,2003:12.(in Chinese)
［15］	Wang Lijun. Research of data sparsity problem in recommendation systems based on collaborative filtering［D］.Changchun:Northeast Normal University,2009:8. (in Chinese)
［16］	Guo Sandang, Wang Lingling. The analysis method of maximum based on degree of grey correlation［J］. Mathematics in Practice and Theory,2013,43(6):195201.(in Chinese)
［17］	Wang Weiping, Wang Jinhui. Hybrid recommendation method based on tag and collaborative filtering［J］. Computer Engineering,2011,37(14):3438.(in Chinese)
	附中文参考文献
［2］	许海玲，吴潇，李晓东,等.互联网推荐系统比较研究［J］. 软件学报，2009，20(2)：350362．
［4］	刘芳先，宋顺林. 改进的协同过滤推荐算法［J］. 计算机工程与应用，2011,47(8):7275．
［5］	韦素云，业宁. 基于项目聚类的全局最近邻的协同过滤算法［J］．计算机科学，2012,39(12):149152．
［6］	李聪，梁昌勇.使用n序访问解析逻辑的协同过滤冷启动消除方法［J］．系统工程理论与实践，2012,32(7):15371545.
［7］	熊忠阳，刘芹. 基于项目分类的协同过滤改进算法［J］．计算机应用研究，2012,29(2):493496.
［8］	吴湖，王永吉，王哲．两阶段联合聚类协同过滤算法［J］．软件学报，2010，21(5):10421054．
［10］	孙小华，陈洪，孔繁胜.在协同过滤中结合奇异值分解与最近邻方法［J］. 计算机应用研究，2006,23(9):206208.
［14］	邓爱琳.电子商务推荐系统关键技术研究［D］．上海：复旦大学，2003:12.
［15］	王立军.基于协同过滤推荐系统的数据稀疏性问题研［D］．长春：东北师范大学，2009:8.
［16］	郭三党，王玲玲．基于最大灰色关联度的聚类方法分析［J］．数学的实践与认识，2013，43(6):195201．
［17］	王卫平,王金辉．基于Tag和协同过滤的混合推荐算法研究［J］．计算机工程，2011,37(14):3438．