File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: Nonnegative low rank matrix approximation for nonnegative matrices

TitleNonnegative low rank matrix approximation for nonnegative matrices
Authors
KeywordsNonnegative matrix
Low-rank approximation
Manifolds
Projections
Issue Date2020
PublisherPergamon. The Journal's web site is located at http://www.elsevier.com/locate/aml
Citation
Applied Mathematics Letters, 2020, v. 105, p. article no. 106300 How to Cite?
AbstractThis paper describes a new algorithm for computing Nonnegative Low Rank Matrix (NLRM) approximation for nonnegative matrices. Our approach is completely different from classical nonnegative matrix factorization (NMF) which has been studied for more than twenty five years. For a given nonnegative matrix, the usual NMF approach is to determine two nonnegative low rank matrices such that the distance between their product and the given nonnegative matrix is as small as possible. However, the proposed NLRM approach is to determine a nonnegative low rank matrix such that the distance between such matrix and the given nonnegative matrix is as small as possible. There are two advantages. (i) The minimized distance by the proposed NLRM method can be smaller than that by the NMF method, and it implies that the proposed NLRM method can obtain a better low rank matrix approximation. (ii) Our low rank matrix admits a matrix singular value decomposition automatically which provides a significant index based on singular values that can be used to identify important singular basis vectors, while this information cannot be obtained in the classical NMF. The proposed NLRM approximation algorithm was derived using the alternating projection on the low rank matrix manifold and the non-negativity property. Experimental results are presented to demonstrate the above mentioned advantages of the proposed NLRM method compared the NMF method.
Persistent Identifierhttp://hdl.handle.net/10722/288096
ISSN
2019 Impact Factor: 3.848
2015 SCImago Journal Rankings: 1.141
ISI Accession Number ID

 

DC FieldValueLanguage
dc.contributor.authorSong, GJ-
dc.contributor.authorNg, MKP-
dc.date.accessioned2020-10-05T12:07:48Z-
dc.date.available2020-10-05T12:07:48Z-
dc.date.issued2020-
dc.identifier.citationApplied Mathematics Letters, 2020, v. 105, p. article no. 106300-
dc.identifier.issn0893-9659-
dc.identifier.urihttp://hdl.handle.net/10722/288096-
dc.description.abstractThis paper describes a new algorithm for computing Nonnegative Low Rank Matrix (NLRM) approximation for nonnegative matrices. Our approach is completely different from classical nonnegative matrix factorization (NMF) which has been studied for more than twenty five years. For a given nonnegative matrix, the usual NMF approach is to determine two nonnegative low rank matrices such that the distance between their product and the given nonnegative matrix is as small as possible. However, the proposed NLRM approach is to determine a nonnegative low rank matrix such that the distance between such matrix and the given nonnegative matrix is as small as possible. There are two advantages. (i) The minimized distance by the proposed NLRM method can be smaller than that by the NMF method, and it implies that the proposed NLRM method can obtain a better low rank matrix approximation. (ii) Our low rank matrix admits a matrix singular value decomposition automatically which provides a significant index based on singular values that can be used to identify important singular basis vectors, while this information cannot be obtained in the classical NMF. The proposed NLRM approximation algorithm was derived using the alternating projection on the low rank matrix manifold and the non-negativity property. Experimental results are presented to demonstrate the above mentioned advantages of the proposed NLRM method compared the NMF method.-
dc.languageeng-
dc.publisherPergamon. The Journal's web site is located at http://www.elsevier.com/locate/aml-
dc.relation.ispartofApplied Mathematics Letters-
dc.subjectNonnegative matrix-
dc.subjectLow-rank approximation-
dc.subjectManifolds-
dc.subjectProjections-
dc.titleNonnegative low rank matrix approximation for nonnegative matrices-
dc.typeArticle-
dc.identifier.emailNg, MKP: michael.ng@hku.hk-
dc.identifier.authorityNg, MKP=rp02578-
dc.description.naturelink_to_subscribed_fulltext-
dc.identifier.doi10.1016/j.aml.2020.106300-
dc.identifier.scopuseid_2-s2.0-85080091000-
dc.identifier.hkuros315733-
dc.identifier.volume105-
dc.identifier.spagearticle no. 106300-
dc.identifier.epagearticle no. 106300-
dc.identifier.isiWOS:000527849300014-
dc.publisher.placeUnited Kingdom-
dc.identifier.issnl0893-9659-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats