个性化微博推荐算法

被引：15
作者：
王晟
王子琪
张铭
机构：
[1] 北京大学信息科学技术学院
来源：
计算机科学与探索 | 2012年 / 6卷 / 10期
关键词：
微博; 推荐; 贝叶斯个性化排序(BPR);
D O I：
暂无
中图分类号：
TP391.3 [检索机];
学科分类号：
081203 ; 0835 ;
摘要：
微博不同于传统的社会网络和电子商务网站,存在用户活跃程度低,微博数据稀疏和用户兴趣动态变化等特点,将传统推荐算法应用于微博推荐时,效果并不理想。提出了一种基于贝叶斯个性化排序的微博推荐算法,对用户进行个性化微博推荐。该基于贝叶斯个性化排序的微博推荐算法,以微博对的形式提取微博系统中的隐式信息,对这些微博对进行学习,从而得到用户对不同微博的兴趣值。根据每条微博发出的时间,估计每条微博对的可信度。发出时间越接近的微博对,它的可信度就越高,并且对用户的兴趣值影响就越大。在新浪微博的真实数据上进行实验和评测,结果表明该基于贝叶斯个性化排序的微博推荐算法相比于对比算法,在进行微博推荐时有更好的效果。
引用
收藏
页码：895 / 902
页数：8
相关论文
共 18 条
[1]  
A collaborative filtering algorithm and evaluation metric that accurately model the user experience. McLaughlin, M,and Herlocker, J. Proceedings of the 27th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval . 2004
[2]  
Collaborative Competitive Filtering:Learning Recommender Using Context of User Choice. Shuang Hong Yang,Bo Long,Alex Smola. Proceedings of the34th international ACM SIGIR conference on Research and development inInformation Retrieval . 2011
[3]  
"Measuring user influence in twitter: the million follower fallacy,". M. Cha,H. Haddadi,F. Benevenuto,K. P. Gummadi. Proc. of International AAAI (Association for the Advancement of Artificial Intelligence) Conf. on Weblogs and Social Media . 2010
[4]  
Finding useful users on Twitter:twittomender the followee recommender. Hannon J,McCarthy K,Smyth B. Proceedings of the33rd European Conference on Information Retrieval (ECIR2011) . 2011
[5]  
Recommending Twitter users to follow using content and collaborative filtering approaches. Hannon J,Bennett M,Smyth B. Proceedings of the4th ACM Conference on Recommender Systems (RecSys10) . 2010
[6]  
Specht G.Using tag recommendations to homogenize folksonomies in microblogging envi-ronments. Zangerle E,Gassler W. Proceedings of the 3rd International Conference on Social Informatics (SocInfo 2011) . 2011
[7]  
Terms of a feature-content-based news recommendation and discovery using Twitter. Phelan O,McCarthy K,Bennett M,et al. Proceedings of the33rd European Conference on Advances in Information Retrieval (ECIR11) . 2011
[8]  
Terms of a feature-content-based news recommendation and discovery using Twitter. Phelan O,McCarthy K,Bennett M,et al. Proceedings of the33rd European Conference on Advances in Information Retrieval (ECIR11) . 2011
[9]  
Short and tweet:experiments on recommending content from information streams. Chen J,Nairn R,Nelson L,et al. Proceedings of the28th International Conference on Human Factors in Computing Systems (CHI10) . 2010
[10]  
The use of the area under the ROC curve in the evaluation of machine learning algorithms. Bradley AP. Pattern Recognition . 1997
12