Multiview clustering via nonnegative matrix factorization based on graph agreement

doi:10.1117/1.JEI.31.4.043024

CORC > 自动化研究所 > 中国科学院自动化研究所 > 智能感知与计算研究中心

	Multiview clustering via nonnegative matrix factorization based on graph agreement
	Zhang, Chengfeng 1; Fu, Wenjun 1; Wang, Guanglong 3; Shi, Lei 3; Meng, Xiangzhu 2
刊名	JOURNAL OF ELECTRONIC IMAGING
	2022-07-01
卷号	31 期号:4 页码:18
关键词	multiview clustering nonnegative matrix factorization graph mechanism
ISSN号	1017-9909
DOI	10.1117/1.JEI.31.4.043024
通讯作者	Meng, Xiangzhu(xiangzhu.meng@cripac.ia.ac.cn)
英文摘要	With the rapid development of information technology, one object can usually be described by multiple views. Intuitively, the diversity and complementarity of different views provide a more comprehensive data description, which can lead to better performance on multiview clustering tasks. Motivated by the fact that how to fully discover diversity and complementary information across views is the key point to deal with multiview clustering, we propose a multiview learning method, termed as graph-agreement nonnegative matrix factorization (GANMF). GANMF attempts to implement this goal based on matrix factorization technology while exploiting the rich multiview information based on the intra- and interview aspects. Specifically, the graph agreement between representation space and raw space is maximized to preserve the intrinsic geometric property in individual view for the intraview case. Similarly, the graph structures in different views are expected to keep consistent with each other by minimizing the divergence between pairwise views. To this end, the intrinsic information and geometric structure information in the intraview case and complementary and compatibility information in the interview case can be simultaneously formulated into one framework. To solve the proposed GANMF, we correspondingly develop an effective algorithm based on iterative alternating strategy. Extensive experimental results on seven multiview datasets demonstrate the superiority and effectiveness of our proposed method.
资助项目	National Natural Science Foundation of PRChina[61672130] ; National Natural Science Foundation of PRChina[72001191]
WOS研究方向	Engineering ; Optics ; Imaging Science & Photographic Technology
语种	英语
出版者	SPIE-SOC PHOTO-OPTICAL INSTRUMENTATION ENGINEERS
WOS记录号	WOS:000848751400042
资助机构	National Natural Science Foundation of PRChina
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/50029]
专题	自动化研究所_智能感知与计算研究中心
通讯作者	Meng, Xiangzhu
作者单位	1.Beijing China Coal Mine Engn Co Ltd, Beijing, Peoples R China 2.Chinese Acad Sci, Ctr Res Intelligent Percept & Comp, Inst Automat, Beijing, Peoples R China 3.Yanchang Petr Barasu Coal Ind Co Ltd, Yulin, Peoples R China
推荐引用方式 GB/T 7714	Zhang, Chengfeng,Fu, Wenjun,Wang, Guanglong,et al. Multiview clustering via nonnegative matrix factorization based on graph agreement[J]. JOURNAL OF ELECTRONIC IMAGING,2022,31(4):18.
APA	Zhang, Chengfeng,Fu, Wenjun,Wang, Guanglong,Shi, Lei,&Meng, Xiangzhu.(2022).Multiview clustering via nonnegative matrix factorization based on graph agreement.JOURNAL OF ELECTRONIC IMAGING,31(4),18.
MLA	Zhang, Chengfeng,et al."Multiview clustering via nonnegative matrix factorization based on graph agreement".JOURNAL OF ELECTRONIC IMAGING 31.4(2022):18.