Max–min distance nonnegative matrix factorization

Handle URI:
http://hdl.handle.net/10754/552386
Title:
Max–min distance nonnegative matrix factorization
Authors:
Wang, Jim Jing-Yan; Gao, Xin ( 0000-0002-7108-3574 )
Abstract:
Nonnegative Matrix Factorization (NMF) has been a popular representation method for pattern classification problems. It tries to decompose a nonnegative matrix of data samples as the product of a nonnegative basis matrix and a nonnegative coefficient matrix. The columns of the coefficient matrix can be used as new representations of these data samples. However, traditional NMF methods ignore class labels of the data samples. In this paper, we propose a novel supervised NMF algorithm to improve the discriminative ability of the new representation by using the class labels. Using the class labels, we separate all the data sample pairs into within-class pairs and between-class pairs. To improve the discriminative ability of the new NMF representations, we propose to minimize the maximum distance of the within-class pairs in the new NMF space, and meanwhile to maximize the minimum distance of the between-class pairs. With this criterion, we construct an objective function and optimize it with regard to basis and coefficient matrices, and slack variables alternatively, resulting in an iterative algorithm. The proposed algorithm is evaluated on three pattern classification problems and experiment results show that it outperforms the state-of-the-art supervised NMF methods.
KAUST Department:
Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division
Citation:
Max–min distance nonnegative matrix factorization 2015, 61:75 Neural Networks
Publisher:
Elsevier BV
Journal:
Neural Networks
Issue Date:
26-Oct-2014
DOI:
10.1016/j.neunet.2014.10.006
Type:
Article
ISSN:
08936080
Additional Links:
http://linkinghub.elsevier.com/retrieve/pii/S0893608014002378
Appears in Collections:
Articles; Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division

Full metadata record

DC FieldValue Language
dc.contributor.authorWang, Jim Jing-Yanen
dc.contributor.authorGao, Xinen
dc.date.accessioned2015-05-06T13:29:32Zen
dc.date.available2015-05-06T13:29:32Zen
dc.date.issued2014-10-26en
dc.identifier.citationMax–min distance nonnegative matrix factorization 2015, 61:75 Neural Networksen
dc.identifier.issn08936080en
dc.identifier.doi10.1016/j.neunet.2014.10.006en
dc.identifier.urihttp://hdl.handle.net/10754/552386en
dc.description.abstractNonnegative Matrix Factorization (NMF) has been a popular representation method for pattern classification problems. It tries to decompose a nonnegative matrix of data samples as the product of a nonnegative basis matrix and a nonnegative coefficient matrix. The columns of the coefficient matrix can be used as new representations of these data samples. However, traditional NMF methods ignore class labels of the data samples. In this paper, we propose a novel supervised NMF algorithm to improve the discriminative ability of the new representation by using the class labels. Using the class labels, we separate all the data sample pairs into within-class pairs and between-class pairs. To improve the discriminative ability of the new NMF representations, we propose to minimize the maximum distance of the within-class pairs in the new NMF space, and meanwhile to maximize the minimum distance of the between-class pairs. With this criterion, we construct an objective function and optimize it with regard to basis and coefficient matrices, and slack variables alternatively, resulting in an iterative algorithm. The proposed algorithm is evaluated on three pattern classification problems and experiment results show that it outperforms the state-of-the-art supervised NMF methods.en
dc.publisherElsevier BVen
dc.relation.urlhttp://linkinghub.elsevier.com/retrieve/pii/S0893608014002378en
dc.rightsArchived with thanks to Neural Networks. http://creativecommons.org/licenses/by-nc-nd/3.0/en
dc.subjectData representationen
dc.subjectNonnegative matrix factorizationen
dc.subjectSupervised learningen
dc.subjectMax–min distance analysisen
dc.titleMax–min distance nonnegative matrix factorizationen
dc.typeArticleen
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Divisionen
dc.identifier.journalNeural Networksen
dc.eprint.versionPublisher's Version/PDFen
kaust.authorWang, Jim Jing-Yanen
kaust.authorGao, Xinen
All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.