File Download
There are no files associated with this item.
Links for fulltext
(May Require Subscription)
- Publisher Website: 10.1016/j.aml.2020.106300
- Scopus: eid_2-s2.0-85080091000
- WOS: WOS:000527849300014
- Find via
Supplementary
- Citations:
- Appears in Collections:
Article: Nonnegative low rank matrix approximation for nonnegative matrices
Title | Nonnegative low rank matrix approximation for nonnegative matrices |
---|---|
Authors | |
Keywords | Nonnegative matrix Low-rank approximation Manifolds Projections |
Issue Date | 2020 |
Publisher | Pergamon. 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? |
Abstract | This 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 Identifier | http://hdl.handle.net/10722/288096 |
ISSN | 2019 Impact Factor: 3.848 2015 SCImago Journal Rankings: 1.141 |
ISI Accession Number ID |
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Song, GJ | - |
dc.contributor.author | Ng, MKP | - |
dc.date.accessioned | 2020-10-05T12:07:48Z | - |
dc.date.available | 2020-10-05T12:07:48Z | - |
dc.date.issued | 2020 | - |
dc.identifier.citation | Applied Mathematics Letters, 2020, v. 105, p. article no. 106300 | - |
dc.identifier.issn | 0893-9659 | - |
dc.identifier.uri | http://hdl.handle.net/10722/288096 | - |
dc.description.abstract | This 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.language | eng | - |
dc.publisher | Pergamon. The Journal's web site is located at http://www.elsevier.com/locate/aml | - |
dc.relation.ispartof | Applied Mathematics Letters | - |
dc.subject | Nonnegative matrix | - |
dc.subject | Low-rank approximation | - |
dc.subject | Manifolds | - |
dc.subject | Projections | - |
dc.title | Nonnegative low rank matrix approximation for nonnegative matrices | - |
dc.type | Article | - |
dc.identifier.email | Ng, MKP: michael.ng@hku.hk | - |
dc.identifier.authority | Ng, MKP=rp02578 | - |
dc.description.nature | link_to_subscribed_fulltext | - |
dc.identifier.doi | 10.1016/j.aml.2020.106300 | - |
dc.identifier.scopus | eid_2-s2.0-85080091000 | - |
dc.identifier.hkuros | 315733 | - |
dc.identifier.volume | 105 | - |
dc.identifier.spage | article no. 106300 | - |
dc.identifier.epage | article no. 106300 | - |
dc.identifier.isi | WOS:000527849300014 | - |
dc.publisher.place | United Kingdom | - |
dc.identifier.issnl | 0893-9659 | - |