A generalized multi-dictionary least squares framework regularized with multi-graph embeddings

doi:10.1016/j.patcog.2019.01.012

	A generalized multi-dictionary least squares framework regularized with multi-graph embeddings
	Abeo, Timothy Apasiba 2,3; Shen, Xiang-Jun 2; Bao, Bing-Kun4 ; Zha, Zheng-Jun 1; Fan, Jianping 5
刊名	PATTERN RECOGNITION
	2019-06-01
卷号	90 页码:1-11
关键词	Multi-view dimension reduction Least squares Multiple graphs Feature extraction Classification
ISSN号	0031-3203
DOI	10.1016/j.patcog.2019.01.012
通讯作者	Shen, Xiang-Jun(xjshen@ujs.edu.cn)
英文摘要	Dimensionality reduction in high dimensional multi-view datasets is an important research topic. It can keep essential features to improve performance in subsequent tasks such as classification and clustering. This paper proposes a generalized framework, which extends the PCA idea of minimizing least squares reconstruction errors, to include data distribution and multiple dictionaries for preserving outliers-free global structures in multi-view datasets. To also preserve local manifold structures, multiple local graphs are incorporated. Finally two models, in Multi-dictionary Least Squares Framework regularized with Multi-graph Embeddings (MD-MGE), are proposed for preserving both global and local structures. Extensive experimental results on four multi-view datasets prove both methods outperform the existing comparative methods. Also, their accuracy rates improvements are statistically significant on all cases below the significance level of 0.05. (C) 2019 Elsevier Ltd. All rights reserved.
资助项目	National Natural Science Foundation of China[61572240] ; National Natural Science Foundation of China[61622211] ; National Natural Science Foundation of China[61472392] ; National Natural Science Foundation of China[61620106009] ; Beijing Natural Science Foundation[4152053]
WOS关键词	CANONICAL CORRELATION-ANALYSIS ; LINEAR DISCRIMINANT-ANALYSIS ; DIMENSIONALITY REDUCTION ; PRESERVING PROJECTIONS ; CLASSIFICATION ; INFORMATION ; EXTENSIONS
WOS研究方向	Computer Science ; Engineering
语种	英语
出版者	ELSEVIER SCI LTD
WOS记录号	WOS:000463130400001
资助机构	National Natural Science Foundation of China ; Beijing Natural Science Foundation
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/28069]
专题	中国科学院自动化研究所
通讯作者	Shen, Xiang-Jun
作者单位	1.Univ Sci & Technol China, Sch Informat Sci & Technol, Hefei, Anhui, Peoples R China 2.JiangSu Univ, Sch Comp Sci & Commun Engn, Zhenjiang 212013, Jiangsu, Peoples R China 3.Tamale Tech Univ, Sch Appl Sci, Box 3ER, Tamale, Ghana 4.Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China 5.Univ North Carolina Charlotte, Dept Comp Sci, Charlotte, NC 28223 USA
推荐引用方式 GB/T 7714	Abeo, Timothy Apasiba,Shen, Xiang-Jun,Bao, Bing-Kun,et al. A generalized multi-dictionary least squares framework regularized with multi-graph embeddings[J]. PATTERN RECOGNITION,2019,90:1-11.
APA	Abeo, Timothy Apasiba,Shen, Xiang-Jun,Bao, Bing-Kun,Zha, Zheng-Jun,&Fan, Jianping.(2019).A generalized multi-dictionary least squares framework regularized with multi-graph embeddings.PATTERN RECOGNITION,90,1-11.
MLA	Abeo, Timothy Apasiba,et al."A generalized multi-dictionary least squares framework regularized with multi-graph embeddings".PATTERN RECOGNITION 90(2019):1-11.