计算机科学 ›› 2014, Vol. 41 ›› Issue (Z6): 394-399.
李贵,陈盛红,李征宇,韩子阳,孙平
LI Gui,CHEN Sheng-hong,LI Zheng-yu,HAN Zi-yang and SUN Ping
摘要: 在推荐系统中,随时间精确捕获用户偏好能有效提高推荐精度。但基于所有用户的简单时间相关性通常是没有实际意义的,因为不同用户的偏好随着外部环境不同而发生改变。用户时下偏好受用户长期偏好和短期偏好的共同影响。为了捕获用户长期和短期偏好,在推荐系统中引入基于会话的时态图STG(Session-based Temporal Graph),提出基于STG的路径融合算法PFA(Path Fusion Algorithm),并生成对某个用户的Top-N物品推荐。使用CiteULike和Delicious两个历史数据集来评估算法的有效性,实验结果表明所 提算法在准确度上要高于以往传统算法。
[1] Ding Y,Li X.Time weight collaborative filtering[C]∥CIKM.2005:485-492 [2] 项亮,陈义,王益.推荐系统实践[M].北京:人民邮电出版社,2012:73-77,1-139 [3] Koren Y.Collaborative filtering with temporal dynamics[C]∥KDD.2009:447-456 [4] Lathia N,Hailes S,Capra L.Temporal collaborative filteringwith adaptive neighbourhoods[C]∥SIGIR.2009:796-797 [5] Ricci F,Nguyen Q N.Acquiring and revising preferences in acritique-based mobile recommender system[J].IEEE Intelligent Systems,2007,22(3):22-29 [6] 王立才,孟祥武,张玉洁.上下文感知推荐系统[J].软件学报,2012,23(1):1-20 [7] Baluja S,Seth R,Sivakumar D,et al.Video suggestion and discovery for youtube:taking random walks through the view graph[C]∥WWW.2008:895-904 [8] Huang Z,Chung W,Chen H.A graph model for e-commerc rec-ommender systems[J].J.Am.Soc.Inf.Sci.Technol.,2004,55(3):259-274 [9] Li Y,Tang J.Expertise search in a time-varying social network[C]∥WAIM’08.2008:293-300 |
No related articles found! |
|