An efficient Nystrom spectral clustering algorithm using incomplete Cholesky decomposition

doi:10.1016/j.eswa.2021.115813

	An efficient Nystrom spectral clustering algorithm using incomplete Cholesky decomposition
	Jia, Hongjie 1,2; Wang, Liangjun 1; Song, Heping 1; Mao, Qirong 1,2; Ding, Shifei 3,4
刊名	EXPERT SYSTEMS WITH APPLICATIONS
	2021-12-30
卷号	186 页码:11
关键词	Spectral clustering Nystrom approximation Incomplete Cholesky decomposition Large data set
ISSN号	0957-4174
DOI	10.1016/j.eswa.2021.115813
英文摘要	Nystrom method can estimate the eigenvectors of a large kernel matrix with the eigenvectors of a small sampled sub-matrix. However, we may encounter two problems when using Nystrom method to speed up spectral clustering: one problem is the approximate eigenvectors generated by standard Nystrom method are sub-optimal, so they may impair the performance of spectral clustering; another one is the accurate Nystrom approximation needs a sufficient number of samples, which will increase the eigen-decomposition cost on the sampled sub-matrix. To solve these problems, this paper proposes an efficient Nystrom spectral clustering algorithm using incomplete Cholesky decomposition, in which a new matrix factorization strategy is designed for Nystrom spectral clustering to meet the orthogonal constraints, and an efficient eigensolver based on incomplete Cholesky decomposition is developed to accelerate the Nystrom approximation. In this way, the obtained approximate orthogonal eigenvectors will help to improve the clustering quality, and the developed eigenvector calculation method will help to reduce the clustering complexity. The experimental results show that the proposed algorithm performs well on many challenging data sets, and it can accomplish more complex clustering tasks with limited computing resources.
资助项目	National Natural Science Foundation of China[61906077] ; National Natural Science Foundation of China[62176106] ; National Natural Science Foundation of China[62172193] ; National Natural Science Foundation of China[61976216] ; National Natural Science Foundation of China[61601202] ; Key Projects of the National Natural Science Foundation of China[U1836220] ; Natural Science Foundation of Jiangsu Province[BK20190838] ; China Postdoctoral Science Foundation[2020M671376] ; China Postdoctoral Science Foundation[2020T130257] ; Natural Science Foundation of the Jiangsu Higher Education Institutions of China[18KJB520009]
WOS研究方向	Computer Science ; Engineering ; Operations Research & Management Science
语种	英语
出版者	PERGAMON-ELSEVIER SCIENCE LTD
WOS记录号	WOS:000705533700007
内容类型	期刊论文
源URL	[http://119.78.100.204/handle/2XEOYT63/17004]
专题	中国科学院计算技术研究所
通讯作者	Mao, Qirong
作者单位	1.Jiangsu Univ, Sch Comp Sci & Commun Engn, Zhenjiang 212013, Jiangsu, Peoples R China 2.Jiangsu Engn Res Ctr Big Data Ubiquitous Percept, Zhenjiang 212013, Jiangsu, Peoples R China 3.China Univ Min & Technol, Sch Comp Sci & Technol, Xuzhou 221116, Jiangsu, Peoples R China 4.Chinese Acad Sci, Inst Comp Technol, Key Lab Intelligent Informat Proc, Beijing 100090, Peoples R China
推荐引用方式 GB/T 7714	Jia, Hongjie,Wang, Liangjun,Song, Heping,et al. An efficient Nystrom spectral clustering algorithm using incomplete Cholesky decomposition[J]. EXPERT SYSTEMS WITH APPLICATIONS,2021,186:11.
APA	Jia, Hongjie,Wang, Liangjun,Song, Heping,Mao, Qirong,&Ding, Shifei.(2021).An efficient Nystrom spectral clustering algorithm using incomplete Cholesky decomposition.EXPERT SYSTEMS WITH APPLICATIONS,186,11.
MLA	Jia, Hongjie,et al."An efficient Nystrom spectral clustering algorithm using incomplete Cholesky decomposition".EXPERT SYSTEMS WITH APPLICATIONS 186(2021):11.