基于超网络中共有邻居的标签相似性CJaccard指标
|
摘要
针对以往社会化标签超网络中仅仅基于局部信息构建标签相似性指标而遇到的局部信息不足或信息稀疏的问题,本文首先阐述了社会化标签超网络的构建方法,然后从全局角度出发,提出基于超网络的标签综合相似度的标签相似性CJaccard指标。选取来自Deli-cious和Last.fm两个社会化标注应用平台的数据集开展谱聚类实验,结果表明,基于超网络的标签综合相似度构建的CJaccard标签相似性指标比基于超网络的标签个体相似度的标签相似性指标具有更高的准确性。
In view of the previous social tagging hypernetwork based on local information only build simi-larity index and tag information is insufficient or local information of sparse problems, this paper first ex-pounds the social tagging method of hypernetwork constructing, and then put forward tag comprehensive simi-larity similarity index of CJaccard based on the hypernetwork from the global perspective. We selected data sets from Delicious and last.fm to carry out spectral clustering experiments The experimental results show that CJaccard tag similarity index based on tag comprehensive similarity in the hypernetwork is more accurate than CJaccard index based on individual similarity in the hypernetwork
In view of the previous social tagging hypernetwork based on local information only build simi-larity index and tag information is insufficient or local information of sparse problems, this paper first ex-pounds the social tagging method of hypernetwork constructing, and then put forward tag comprehensive simi-larity similarity index of CJaccard based on the hypernetwork from the global perspective. We selected data sets from Delicious and last.fm to carry out spectral clustering experiments The experimental results show that CJaccard tag similarity index based on tag comprehensive similarity in the hypernetwork is more accurate than CJaccard index based on individual similarity in the hypernetwork
引文
[1]蒋盛益等.国内外社会化标签挖掘研究综述.图书情报工作,2014.58(21):136-145.
[2]李慧宗,胡学钢,何伟,潘剑寒.社会化标注环境下的标签共现谱聚类方法[J].图书情报工作,2014,58(23):129-135.
[3]李慧宗等.基于LDA的社会化标签综合聚类方法[J].情报学报,2015.34(2):146-155.
[4]Fan,Z.,et al.,A Text Clustering Approach of Chinese News Based on Neural Network Language Model.International Journal of Parallel Programming,2016.44(1):198-206.
[5]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(05):444-451.
[6]Rawashdeh M,Alhamid M F,Kim H N,et al.Graph-base d personalized recommendation in social tagging systems[C].2014 IEEE International Conference on Multimedia and Expo Workshops(ICMEW)IEEE Computer Society,2014:1-6.
[7]李涛.基于超网络的社会化标签相似性度量研究[D].杭州:浙江理工大学,2016.
[8]郭进利,祝昕昀.超网络中标度律的涌现[J].物理学报,2014,63(9):90207-090207.
[9]Kaufman,L.and P.J.Rousseeuw,Finding Groups in Data:An Introduction to Cluster Analysis,1990:DBLP.
[2]李慧宗,胡学钢,何伟,潘剑寒.社会化标注环境下的标签共现谱聚类方法[J].图书情报工作,2014,58(23):129-135.
[3]李慧宗等.基于LDA的社会化标签综合聚类方法[J].情报学报,2015.34(2):146-155.
[4]Fan,Z.,et al.,A Text Clustering Approach of Chinese News Based on Neural Network Language Model.International Journal of Parallel Programming,2016.44(1):198-206.
[5]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(05):444-451.
[6]Rawashdeh M,Alhamid M F,Kim H N,et al.Graph-base d personalized recommendation in social tagging systems[C].2014 IEEE International Conference on Multimedia and Expo Workshops(ICMEW)IEEE Computer Society,2014:1-6.
[7]李涛.基于超网络的社会化标签相似性度量研究[D].杭州:浙江理工大学,2016.
[8]郭进利,祝昕昀.超网络中标度律的涌现[J].物理学报,2014,63(9):90207-090207.
[9]Kaufman,L.and P.J.Rousseeuw,Finding Groups in Data:An Introduction to Cluster Analysis,1990:DBLP.