A self-supervised framework for clustering ensemble

ISCAS OpenIR

	A self-supervised framework for clustering ensemble
	Du, Liang (1); Shen, Yi-Dong (1); Shen, Zhiyong (4); Wang, Jianying (5); Xu, Zhiwu (1)
	2013
会议名称	14th International Conference on Web-Age Information Management, WAIM 2013
页码	253-264
会议日期	June 14, 2013 - June 16, 2013
会议地点	Beidaihe, China
收录类别	EI
出版地	Springer Verlag, Tiergartenstrasse 17, Heidelberg, D-69121, Germany
ISSN	3029743
ISBN	9783642385612
部门归属	(1) State Key Laboratory of Computer Science, Institute of Software, Chinese Academy of Sciences, Beijing 100190, China; (2) Graduate University of Chinese Academy of Sciences, China; (3) University of Chinese Academy of Sciences, Beijing 100049, China; (4) Baidu Inc., Beijing 100085, China; (5) Computing Center, Shanghai University, Shanghai, China
摘要	Clustering ensemble refers to combine a number of base clusterings for a particular data set into a consensus clustering solution. In this paper, we propose a novel self-supervised learning framework for clustering ensemble. Specifically, we treat the base clusterings as pseudo class labels and learn classifiers for each of them. By adding priors to the parameters of these classifiers, we capture the relationships between different base clusterings and meanwhile obtain a a single consolidated clustering result. In the proposed framework, we are able to incorporate the original data features to improve the performance of clustering ensemble. Another advantage, which distinguishes the proposed framework from the traditional clustering ensemble approaches, is with the generalization capability, i.e. it is able to assign the incoming data instances to the consensus clusters directly based on the original data features. We conduct extensive experiments on multiple real world data sets to show the effectiveness of our method. © 2013 Springer-Verlag Berlin Heidelberg.; Clustering ensemble refers to combine a number of base clusterings for a particular data set into a consensus clustering solution. In this paper, we propose a novel self-supervised learning framework for clustering ensemble. Specifically, we treat the base clusterings as pseudo class labels and learn classifiers for each of them. By adding priors to the parameters of these classifiers, we capture the relationships between different base clusterings and meanwhile obtain a a single consolidated clustering result. In the proposed framework, we are able to incorporate the original data features to improve the performance of clustering ensemble. Another advantage, which distinguishes the proposed framework from the traditional clustering ensemble approaches, is with the generalization capability, i.e. it is able to assign the incoming data instances to the consensus clusters directly based on the original data features. We conduct extensive experiments on multiple real world data sets to show the effectiveness of our method. © 2013 Springer-Verlag Berlin Heidelberg.
语种	英语
内容类型	会议论文
URI标识	http://ir.iscas.ac.cn/handle/311060/16679
专题	中国科学院软件研究所
推荐引用方式 GB/T 7714	Du, Liang ,Shen, Yi-Dong ,Shen, Zhiyong ,et al. A self-supervised framework for clustering ensemble[C]. Springer Verlag, Tiergartenstrasse 17, Heidelberg, D-69121, Germany,2013:253-264.