基于概率矩阵分解的多指标协同过滤算法

doi:10.6040/j.issn.1672-3961.2.2015.080

山东大学学报(工学版) ›› 2016, Vol. 46 ›› Issue (3): 65-73.doi: 10.6040/j.issn.1672-3961.2.2015.080

基于概率矩阵分解的多指标协同过滤算法

庞俊涛¹, 张晖^2*, 杨春明¹, 李波^1,3, 赵旭剑¹

1. 西南科技大学计算机科学与技术学院, 四川绵阳 621010;2. 西南科技大学教育信息化推进办公室, 四川绵阳 621010;3. 中国科学技术大学计算机科学与技术学院, 安徽合肥 230026

收稿日期:2015-06-23 出版日期:2016-06-30 发布日期:2015-06-23
通讯作者: 张晖(1972— ),男, 安徽宿松人,教授,工学博士,主要研究方向为文本挖掘与知识工程.E-mail:zhanghui@swust.edu.cn E-mail:pangjuntaoer@163.com
作者简介:庞俊涛(1989— ),男,四川通江人,硕士研究生,主要研究方向为机器学习与推荐系统.E-mail:pangjuntaoer@163.com
基金资助:
四川省教育厅资助项目(14ZB0113);西南科技大学博士基金资助项目(12zx7116)

Multi-criteria collaborative filtering algorithm based on probabilistic matrix factorization

PANG Juntao¹, ZHANG Hui^2*, YANG Chunming¹, LI Bo^1,3, ZHAO Xujian¹

1.School of Computer Science and Technology, Southwest University of Science and Technology, Mianyang 621010, Sichuan, China;
2. Educational Informationization Office, Southwest University of Science and Technology, Mianyang 621010, Sichuan, China;
3. School of Computer Science and Technology, University of Science and Technology of China, Hefei 230026, Anhui, China

Received:2015-06-23 Online:2016-06-30 Published:2015-06-23

摘要/Abstract

摘要： 为解决已有关于多指标评分推荐方法中忽略多指标之间存在相关性的问题,提出一种基于概率矩阵分解的多指标协同过滤算法(multi-criteria collaborative filtering algorithm based on probabilistic matrix factorization, MCPMF)。该算法将多指标评分表示成一个对整体用户和产品产生影响的权重矩阵,并假设该矩阵潜在分布服从高斯分布,其概率密度分布与用户和产品特征矩阵的概率密度分布条件相关。通过概率矩阵分解的方法学习得到用户和产品特征矩阵。在两个真实数据集上的试验结果表明,该方法比只考虑单一综合评分的方法能更加精确地预测用户的综合评分,同时能降低数据稀疏对推荐算法的影响。

关键词: 推荐系统, 协同过滤, 概率矩阵分解, 多指标

Abstract: In order to solve the problem that the correlation was neglected among the multi-criteria in the recommendation method of the multi-criteria, a multi-criteria collaborative filtering algorithm based on probabilistic matrix factorization(MCPMF)was proposed. The algorithm represented the multi-criteria as a weight-matrix which has influence on all users and items. The latent distribution of the weight-matrix was assumed to follow Gaussian distribution, and the probability density distribution of the matrix was conditional related to the distribution of user and item latent feature matrix. The user and item feature matrix was learned by probability matrix factorization method. Experimental results on two real datasets showed that the proposed method was more accurate in forecasting the user's overall rating compared with methods which only considered single overall rating and could reduce the impact of data sparsity to recommendation algorithms.

Key words: probabilistic matrix factorizetion, collaborative filtering, recommendation system, multi-criteria

中图分类号:

TP181

庞俊涛, 张晖, 杨春明, 李波, 赵旭剑. 基于概率矩阵分解的多指标协同过滤算法[J]. 山东大学学报(工学版), 2016, 46(3): 65-73.

PANG Juntao, ZHANG Hui, YANG Chunming, LI Bo, ZHAO Xujian. Multi-criteria collaborative filtering algorithm based on probabilistic matrix factorization[J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2016, 46(3): 65-73.

参考文献

[1] ADOMAVICIUS G, TUZHILIN A. Toward the next generation of recommender systems: a survey of the state-of-the-art and possible extensions[J]. Knowledge and Data Engineering, IEEE Transactions on, 2005, 17(6):734-749.
[2] LAKIOTAKI K, TSAFARAKIS S, MATSATSINIS N. UTA-Rec: a recommender system based on multiple criteria analysis[C] //Proceedings of the 2008 ACM Conference on Recommender Systems. Lausann, Switzerland: ACM, 2008:219-226.
[3] MANOUSELIS N, COSTOPOULOU C. Analysis and classification of multi-criteria recommender systems[J]. World Wide Web, 2007, 10(4):415-441.
[4] MANOUSELIS N, COSTOPOULOU C. Experimental analysis of design choices in multiattribute utility collaborative filtering[J]. International Journal of Pattern Recognition and Artificial Intelligence, 2007, 21(2):311-331.
[5] ADOMAVICIUS G, KWON Y O. New recommendation techniques for multicriteria rating systems[J]. Intelligent Systems, IEEE, 2007, 22(3):48-55.
[6] JANNACH D, KARAKAYA Z, GEDIKLI F. Accuracy improvements for multi-criteria recommender systems[C] //Proceedings of the 13th ACM Conference on Electronic Commerce. Valencia, Spain: ACM, 2012:674-689.
[7] HWANG C S. Genetic algorithms for feature weighting in multi-criteria recommender systems[J].Journal of Convergence Information Technology, 2010(5):126-136.
[8] TSOUKIàs A, MATSATSINIS N, LAKIOTAKI K. Multi-criteria user modeling in recommender systems[J]. IEEE Intelligent Systems, 2011, 26(2):64-76.
[9] LAKIOTAKI K, TSAFARAKIS S, MATSATSINIS N. UTA-Rec: a recommender system based on multiple criteria analysis[C] //Proceedings of the 2008 ACM Conference on Recommender Systems. Lausann, Switzerland: ACM, 2008:219-226.
[10] MANOUSELIS N, COSTOPOULOU C. Analysis and classification of multi-criteria recommender systems[J]. World Wide Web, 2007, 10(4):415-441.
[11] NILASHI M, JANNACH D, IBRAHIM O, et al. Clustering-and regression-based multi-criteria collaborative filtering with incremental updates[J]. Information Sciences, 2015, 293(293):235-250.
[12] NILASHI M, IBRAHIM O, ITHNIN N. Multi-criteria collaborative filtering with high accuracy using higher order singular value decomposition and Neuro-Fuzzy system[J]. Knowledge-Based Systems, 2014, 60(2):82-101.
[13] NILASHI M, IBRAHIM O B, ITHNIN N, et al. A multi-criteria recommendation system using dimensionality reduction and Neuro-Fuzzy techniques[J]. Soft Computing, 2015, 19(11):3173-3207.
[14] 张付志, 常俊风, 王栋. 基于 Widrow-Hoff 神经网络的多指标推荐算法[J]. 模式识别与人工智能, 2011, 24(2):233-242. ZHANG Fuzhi, CHANG Junfeng,WANG Dong. Multi-riteria recommendation algorithm based on widrow-hoff neural network[J]. Pattern Recognition and Artificial Intelligence, 2011, 24(2):233-242.
[15] ZHANG Y, ZHUANG Y, WU J, et al. Applying probabilistic latent semantic analysis to multi-criteria recommender system[J]. Ai Communications, 2009, 22(2): 97-107.
[16] DUECK D, FREY B, DUECK D, et al. Probabilistic sparse matrix factorization[EB/OL].(2004-09-28)[2014-04-21]. http://www.researchgate.net/publication/240191894_-Probabilistic_-Sparse_-Matrix_-Factorization.
[17] MNIH A, SALAKHUTDINOV R. Probabilistic matrix factorization[C] //Advances in Neural Information Processing Systems. Vancouver, Canada: MIT Press, 2007:1257-1264.
[18] SALAKHUTDINOV R, MNIH A. Bayesian probabilistic matrix factorization using Markov chain Monte Carlo[C] //Proceedings of the 25th International Conference on Machine Learning. Helsinki, Finland: ACM, 2008: 880-887.
[19] ZHOU T, SHAN H, BANERJEE A, et al. Kernelized probabilistic matrix factorization: exploiting graphs and side information[C]. SDM 2012.California, USA: SDM, 2012, 12:403-414.
[20] LI Z, LIU J, ZHU X, et al. Image annotation using multi-correlation probabilistic matrix factorization[C] //Proceedings of the International Conference on Multimedia. Firenze, Italia: ACM, 2010:1187-1190.
[21] PATEREK A. Improving regularized singular value decomposition for collaborative filtering[C] //Proceedings of KDD Cup and Workshop. California, USA: ACM, 2007, 2007: 5-8.
[22] FUCHS M, ZANKER M. Multi-criteria Ratings for Recommender Systems: An Empirical Analysis in the Tourism Domain[J]. Lecture Notes in Business Information Processing, 2012, 123:100-111.

多维度评价

Viewed

Full text

Abstract

Cited

Shared

Discussed

基于概率矩阵分解的多指标协同过滤算法

Multi-criteria collaborative filtering algorithm based on probabilistic matrix factorization

PDF (PC)

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

多维度评价

本文评价

推荐阅读 10

[1]	段圣宇,吴伊宁,赛高乐. 一种面向矩阵分解模型的推荐系统训练加速方法[J]. 山东大学学报 (工学版), 2025, 55(1): 24-29.
[2]	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.
[3]	郑顺,王绍卿,刘玉芳,李可可,孙福振. 基于动态掩码和多对对比学习的序列推荐模型[J]. 山东大学学报 (工学版), 2023, 53(6): 47-55.
[4]	王冰,马文明,武聪,郝昱猛. 融合信任相似度的偏置概率矩阵分解算法[J]. 山东大学学报 (工学版), 2022, 52(4): 110-117.
[5]	张胜男,王雷,常春红,郝本利. 基于三维剪切波变换和BM4D的图像去噪方法[J]. 山东大学学报 (工学版), 2020, 50(2): 83-90.
[6]	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.
[7]	胡云,张舒,李慧,佘侃侃,施珺. 基于信任网络重构的推荐算法[J]. 山东大学学报 (工学版), 2019, 49(2): 42-46.
[8]	黄丹,王志海,刘海洋. 一种局部协同过滤的排名推荐算法[J]. 山东大学学报(工学版), 2016, 46(5): 29-36.
[9]	林耀进,张佳,林梦雷,王娟. 一种基于模糊信息熵的协同过滤推荐方法[J]. 山东大学学报(工学版), 2016, 46(5): 13-20.
[10]	李朔,石宇良. 基于位置社交网络中地点聚类推荐方法[J]. 山东大学学报(工学版), 2016, 46(3): 44-50.
[11]	张佳,林耀进,林梦雷,刘景华,李慧宗. 基于信息熵的协同过滤算法[J]. 山东大学学报(工学版), 2016, 46(2): 43-50.
[12]	孙远帅,陈垚,刘向荣,陈珂,林琛. 基于项目层次相似性的推荐算法[J]. 山东大学学报(工学版), 2014, 44(3): 8-14.
[13]	陈大伟,闫昭*,刘昊岩. SVD系列算法在评分预测中的过拟合现象[J]. 山东大学学报(工学版), 2014, 44(3): 15-21.
[14]	李改1,2,3, 李磊2,3. 一种解决协同过滤系统冷启动问题的新算法[J]. 山东大学学报(工学版), 2012, 42(2): 11-17.
[15]	王爱国,李廉*,杨静,陈桂林. 一种基于Bayesian网络的网页推荐算法[J]. 山东大学学报(工学版), 2011, 41(4): 137-142.