一种基于混合粒度的微博用户标签推荐模型

doi:10.11896/j.issn.1002-137X.2016.04.039

摘要/Abstract

摘要： 针对已有的标签推荐模型在实际微博场景运用中存在的多样性、相关性较差等不足,提出了一种基于混合粒度的标签推荐模型。将微博用户的可分析资源分解成由用户信息、标签和微博正文组成的混合粒度,在不同粒度上分别进行个人信息过滤及个性标签分析,从而计算用户标签的熵值与内联度和分类标注标签词汇,提取微博正文主题等,最终为用户推荐具有较强关联性的个性化标签。与一般LDA模型的对比实验证明,该模型可以有效解决新用户的冷启动、标签推荐的准确度等问题,同时保证了推荐的多样性。

关键词: 社会化标签,混合粒度,主题提取,社交网络,多样性

Abstract: To avoid the shortcomings of existing tag-recommendation system,we proposed a recommendation model based on hybrid grain.The proposed model divides the resources to multiple grain including user information,tags and blog content.Then filtering personal information,analyzing personality tags,calculating user tag entropy and inline degree,classification words,and extracting topic of micro-blog can proceed based on the divided hybrid grain.Through the steps above,the proposed method can recommend associated personalized labels to users.By comparing the experimental results,it shows that the proposed model can efficiently deal with the cold boot problem,improve the accuracy of tag recommendation and insure the diversity of recommended labels.

Key words: Social tagging,Hybrid grain,Topic extraction,Social network,Diversity

张瑞,金志刚,王颖. 一种基于混合粒度的微博用户标签推荐模型[J]. 计算机科学, 2016, 43(4): 192-196. https://doi.org/10.11896/j.issn.1002-137X.2016.04.039

ZHANG Rui, JIN Zhi-gang and WANG Ying. Recommendation Model of Microblog User Tags Based on Hybrid Grain[J]. Computer Science, 2016, 43(4): 192-196. https://doi.org/10.11896/j.issn.1002-137X.2016.04.039

参考文献

[1] Di L,Du Y P.Application of LDA Model in Microblog UserRecommendation[J].Computer Engineering,2014(5):1-6,11(in Chinese) 邸亮,杜永萍.LDA模型在微博用户推荐中的应用[J].计算机工程,2014(5):1-6,11
[2] Ding Z Y,Jia Y,Zhou B,et al.Survey of Influence Analysis for Social Networks[J].Computer Science,2014,1(1):48-53(in Chinese) 丁兆云,贾焰,周斌,等.社交网络影响力研究综述[J].计算机科学,2014,1(1):48-53
[3] Denning P J.Computing is a natural science[J].Communications of the ACM,2007,50(7):13-18
[4] Jiang W Q.Zipf and the Principle of Least Effort[J].Tongji University Journal(Social Science Section),2005,16(1):87-95(in Chinese) 姜望琪.Zipf与省力原则[J].同济大学学报(社会科学版),2005,16(1):87-95
[5] Sun D T,He T,Zhang F H.Survey of Cold-start Problem in Collaborative Filtering Recommender System[J].Computer and Modernization,2012(5):59-63(in Chinese) 孙冬婷,何涛,张福海.推荐系统中的冷启动问题研究综述[J].计算机与现代化,2012(5):59-63
[6] Zhang Z K,Zhou T,Zhang Y C.Tag-aware recommender systems:a state-of-the-art survey[J].Journal of Computer Science and Technology,2011,26(5):767-777
[7] Guan Z,Bu J,Mei Q.Personalized tag recommendation using graph-based ranking on multi-type interrelated objects[C]∥Proceedings of the 32nd International ACM SIGIR Conference on Research and Development in Information Retrieval.ACM,2009:540-547
[8] Zhang N,Zhang Y,Tang J.A tag recommendation system based on contents[C]∥Proceedings of the ECML PKDD Discovery Challenge Workshop,2009(DC09).2009:285
[9] Xu B,Yang D,Zhang Y,et al.Relationship Bind Topic Model Toward Tag Recommendation for Micro-Blog Users[J].Journal of Frontiers of Computer Science and Technology,2014(3):288-295(in Chinese) 徐彬,杨丹,张昱,等.面向微博用户标签推荐的关系约束主题模型[J].计算机科学与探索,2014(3):288-295
[10] Chen Y,Lin L,Sun C J,et al.A Tag Recommendation Method for Microblog Users[J].Intelligent Computer and Applications,2011(5):21-26(in Chinese) 陈渊,林磊,孙承杰,等.一种面向微博用户的标签推荐方法[J].智能计算机与应用,2011(5):21-26
[11] Hu J,Wang B,Liu Y.Personalized tag recommendation usingsocial influence[J].Journal of Computer Science and Technology,2012,27(3):527-540
[12] Cai M S,Li X M,Yin Y T.Hybrid top-N recommendation method based on social user tag [J].Application Research of Computers,2013(5):1309-1311,1344(in Chinese) 蔡孟松,李学明,尹衍腾.基于社交用户标签的混合top-N推荐方法[J].计算机应用研究,2013(5):1309-1311,1344
[13] Liao Z F,Wang C Q,Li X Q,et al.Tag Recommendation and New User Tag Recommendation Algorithms Based on Tensor Decomposition [J].Journal of Chinese Computer Systems,2013,34(11):2472-2476(in Chinese) 廖志芳,王超群,李小庆,等.张量分解的标签推荐及新用户标签推荐算法[J].小型微型计算机系统,2013,34(11):2472-2476
[14] Wang S,Zhang L M.Mining Algorithm and Structural Analysis of Microblog Interpersonal Relationship Network Based on Tag [J].Computer Engineering,2014,40(5):7-11(in Chinese) 王莎,张连明.基于标签的微博人脉网络挖掘算法和结构分析[J].计算机工程,2014,40(5):7-11
[15] Atallah M J.Algorithms and Theory of Computation Handbook[M].CRC Press,1998
[16] Chen L F,Mark-Liao H Y,Ko M T,et al.A new LDA-based face recognition system which can solve the small sample size problem[J].Pattern Recognition,2000,3:1713-1726
[17] Heinrich G.Parameter estimation for text analysis[R].2004
[18] Liu Yang,Qiu Ming-hui,Gottipati S,et al.CQARank:Jointly Model Topics and Expertise in Community Question Answering[C]∥Proceedings of the 22nd ACM International Conference on Information and Knowledge Management (CIKM 2013).2013:99-108
[19] 定型数据集和测试数据集.http://msdn.microsoft.com/zh-cn/library/bb895173.aspx

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed