基于张量分解的个性化信息推荐方法优化研究
|
摘要
个性化信息标志着用户从被动的信息消费者变为主动的信息创造者,而个性化信息推荐是通过用户以往的标记行为,向用户提供将来可能用到的信息。这意味着向每一个用户推荐一张个性化的信息清单,这个清单既取决于资源的内容也取决于用户本身。针对已有的基于张量分解中的塔克分解的信息推荐的缺点——三维立方核心张量导致的个性化信息推荐的立方运行时间加大,尤其是大的数据集,提出新的张量分解方法来缩短系统运行时间,最后进行方法的优化验证。
Personalized information marks that user becomes from passive consumers into active information creator, and personalized information recommendation provides tags to the users in the future to help mark process through the user's previous mark behavior. This means that each user will be recommended a personalized labels list, and the recommended list of tags depends on both resources and the user. Then on the tensor decomposition-tucker's comings-running time increasing led dimensional cubic core tensor, especially large data sets, it puts forward the new tensor decomposition method to shorten the system running time,finally take the optimization of verification.
Personalized information marks that user becomes from passive consumers into active information creator, and personalized information recommendation provides tags to the users in the future to help mark process through the user's previous mark behavior. This means that each user will be recommended a personalized labels list, and the recommended list of tags depends on both resources and the user. Then on the tensor decomposition-tucker's comings-running time increasing led dimensional cubic core tensor, especially large data sets, it puts forward the new tensor decomposition method to shorten the system running time,finally take the optimization of verification.
引文
1 GEMMELL J,etal.Personalizing navigation in folksonomies using hierarchical tag clustering[EB/OL].http://dl.acm.org/citation.cfm?id=1430581,2012-09-18.
2 Deerwester S,Dumais S T,Furnas G W,etal.Indexing by latent semantic analysi[EB/OL].http://www.cob.unt.edu/itds/facul ty/evangelopoulos/dsci5910/LSA_Deerwester1990.pdf,2012-08-18.
3 Hofmann T.Probabilistic latent semantic ind exing[EB/OL].http://lvk.cs.msu.su/~bruzz/articles/feature_Selection_clustering/Hofma nn.SIGIR99.pdf,2012-08-18.
4 Blei D M,Ng A Y,Jordan M I.Latent dirichlet allocation[EB/OL].http://admis.fu dan.edu.cn/seminars/ppt/lecture-lda.pdf,2012-08-18.
5 SHIRATSUCHI K,etal.Finding unknown intere-sts utilizing the wisdom of crowds in a social bookmark service[C]//Proceedings of the 2006 IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Tech-nology.Washington:IEEE Computer Society,2006:421-424.
6 田宗若.张量分析[M].西安:西北工业大学出版社,2005:78.
7 CICHOCKIA,ZDUNEK R,PHAN A H,et al.Nonnegativematrix and tensor factorizations:applications to exploratory multi-way data analy-sis and blind source separation[M].Hobolcen:W iley,2009:47.
8 KOLDA TG,BADER BW.Tensor decompositi on ns and applications[J].S IAM Review,2009,51(3):455-500.
9 KROONENBERG PM.Appliedmultiway data analysis[M].Hoboken:Wiley,2008:164.
10 S.Rendle,L.BalbyMarinho,A.NanoPou los,and L.Sehmidt-Thieme.Leaming Optimal ranking with tensor faetorization for tag recommendation[C].In Proeee-dings of the 15th ACM SIGKDD intem-ational conference on Kowledge diseov-ery and data mining ACM,2009:727一736.
11 张子柯.社会化标签系统的结构、演化和功能[J].上海理工大学学报,2011,33(5):445-451.
12 夏建平,周激流,何坤,刘褪,王东方.基于小波变换和Tucke:分解的彩色图像数字水印[J].计算机应用研究,2010,27(2):798-800.
13 蒋红卫.一种新的多向数据分析方法-多重双线性分解及运用初步[D].上海:第二军医大学,2010.
2 Deerwester S,Dumais S T,Furnas G W,etal.Indexing by latent semantic analysi[EB/OL].http://www.cob.unt.edu/itds/facul ty/evangelopoulos/dsci5910/LSA_Deerwester1990.pdf,2012-08-18.
3 Hofmann T.Probabilistic latent semantic ind exing[EB/OL].http://lvk.cs.msu.su/~bruzz/articles/feature_Selection_clustering/Hofma nn.SIGIR99.pdf,2012-08-18.
4 Blei D M,Ng A Y,Jordan M I.Latent dirichlet allocation[EB/OL].http://admis.fu dan.edu.cn/seminars/ppt/lecture-lda.pdf,2012-08-18.
5 SHIRATSUCHI K,etal.Finding unknown intere-sts utilizing the wisdom of crowds in a social bookmark service[C]//Proceedings of the 2006 IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Tech-nology.Washington:IEEE Computer Society,2006:421-424.
6 田宗若.张量分析[M].西安:西北工业大学出版社,2005:78.
7 CICHOCKIA,ZDUNEK R,PHAN A H,et al.Nonnegativematrix and tensor factorizations:applications to exploratory multi-way data analy-sis and blind source separation[M].Hobolcen:W iley,2009:47.
8 KOLDA TG,BADER BW.Tensor decompositi on ns and applications[J].S IAM Review,2009,51(3):455-500.
9 KROONENBERG PM.Appliedmultiway data analysis[M].Hoboken:Wiley,2008:164.
10 S.Rendle,L.BalbyMarinho,A.NanoPou los,and L.Sehmidt-Thieme.Leaming Optimal ranking with tensor faetorization for tag recommendation[C].In Proeee-dings of the 15th ACM SIGKDD intem-ational conference on Kowledge diseov-ery and data mining ACM,2009:727一736.
11 张子柯.社会化标签系统的结构、演化和功能[J].上海理工大学学报,2011,33(5):445-451.
12 夏建平,周激流,何坤,刘褪,王东方.基于小波变换和Tucke:分解的彩色图像数字水印[J].计算机应用研究,2010,27(2):798-800.
13 蒋红卫.一种新的多向数据分析方法-多重双线性分解及运用初步[D].上海:第二军医大学,2010.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |