推荐算法综述

被引:85
作者
杨博 [1 ,2 ]
赵鹏飞 [1 ,2 ]
机构
[1] 吉林大学计算机科学与技术学院
[2] 吉林大学符号计算与知识工程教育部重点实验室
关键词
信息过载; 推荐系统; 协同过滤; 信息检索; 数据挖掘; 机器学习;
D O I
暂无
中图分类号
TP301.6 [算法理论];
学科分类号
081202 ;
摘要
推荐是解决互联网信息过载的主要途径之一,已被广泛应用于电子商务等多个领域.尽管已存在多种推荐算法,建造出更加智能、更加鲁棒的推荐系统仍面临诸多尚未解决的难题,推荐方法的研究仍是智能信息处理的研究热点.文章首先阐述了推荐方法的研究背景、研究意义,之后分别介绍了协同过滤推荐算法、基于内容的推荐算法、基于图结构的推荐算法和混合推荐算法,分析了各类算法的优点与不足,最后总结了主要的评价方法以及面临的主要问题,提出了改进的方法和未来可能的研究方向.
引用
收藏
页码:337 / 350
页数:14
相关论文
共 64 条
[1]  
一种探测推荐系统托攻击的无监督算法[J]. 李聪,骆志刚,石金龙. 自动化学报. 2011(02)
[2]  
Collaborative Filtering with Privacy Via Factor Analysis. Canny J. Process of the 25th International ACM SIGIRConference on Research and Development in Information Retrieval . 2002
[3]  
Amazon.com recommendations: item-to-item collaborative filtering. Greg Linden, Brent Smith, Jeremy York. IEEE Internet Computing . 2003
[4]  
Personal recommendation based on weighted bipartite networks. Liu Jie,Shang MS,Chen D B. The6thInternational Con-ference on Fuzzy Systems and Knowledge Discovery . 2009
[5]  
Using probabilistic relational models for collaborative filtering. Getoor L,Sahami M. Proc. of the Workshop on Web Usage Analysis and User Profiling under KDD’99 (WEBKDD’99) . 1999
[6]  
Time weight collaborative filtering. Ding Y,Li X. Proceedings of the 14th ACM International Conference on Infor-mation and Knowledge Management . 2005
[7]  
Semiotic dynamics and collaborative tagging. Cattuto C,Loreto V,Pietronero L. Proceedings of the National Academy of Sciences of the United States of America . 2007
[8]  
Slope one Predictors for Online Rating-based Collaborative Filtering. Lemire D,Maclachlan A. Process of the SIAMData Mining Conference . 2005
[9]  
Knowledge-Based recommender systems. Burke R. Encyclopedia of Library and Information Systems . 2000
[10]  
Collaborative Filtering withDecoupled Models for Preferences and Rratings. Jin R,Si L,Zhai C,et al. Proceedings of the 12th International Conference on In-formation and Knowledge Management . 2003