A nonnegative projection based algorithm for low-rank nonnegative matrix approximation

Peitao Wang; Zhaoshui He; Kan Xie; Junbin Gao; Michael Antolovich

doi:10.1007/978-3-319-70087-8_26

A nonnegative projection based algorithm for low-rank nonnegative matrix approximation

Peitao Wang, Zhaoshui He, Kan Xie, Junbin Gao, Michael Antolovich

Guangdong University of Technology

Research output: Book chapter/Published conference paper › Conference paper › peer-review

1 Citation (Scopus)

Abstract

Nonnegative matrix factorization/approximation (NMF/NMA) is a widely used method for data analysis. So far, many multiplicative update algorithms have been developed for NMF. In this paper, we propose a nonnegative projection based NMF algorithm, which is different from the conventional multiplicative update NMF algorithms and decreases the objective function by performing Procrustes rotation and nonnegative projection alternately. The experiment results demonstrate that the new algorithm converges much faster than traditional ones.

Original language	English
Title of host publication	Neural information processing
Subtitle of host publication	24th International Conference, ICONIP 2017, Guangzhou, China, November 14-18, 2017, Proceedings, Part I
Editors	Derong Liu, Shengli Xie, Yuanqing Li, Dongbin Zhao, El-Sayed M. El-Alfy
Publisher	Springer-Verlag London Ltd.
Pages	240-247
Number of pages	8
ISBN (Electronic)	9783319700878
ISBN (Print)	9783319700861
DOIs	https://doi.org/10.1007/978-3-319-70087-8_26
Publication status	Published - 24 Oct 2017
Event	24th International Conference on Neural Information Processing: ICONIP 2017 - Ramada Pearl Hotel, Guangzhou, China Duration: 14 Nov 2017 → 18 Nov 2017 http://www.conference123.org/iconip2017/index.html (Conference website) https://link.springer.com/book/10.1007/978-3-319-70087-8#about (Conference proceedings)

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	10634 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	24th International Conference on Neural Information Processing
Country/Territory	China
City	Guangzhou
Period	14/11/17 → 18/11/17
Other	The 24th International Conference on Neural Information Processing will be held in Guangzhou, China, November 14–18, 2017. It is an annual event, organized since 1994 by the Asia Pacific Neural Network Society (APNNS, previously APNNA).
Internet address	http://www.conference123.org/iconip2017/index.html (Conference website) https://link.springer.com/book/10.1007/978-3-319-70087-8#about (Conference proceedings)

Access to Document

10.1007/978-3-319-70087-8_26

Cite this

Wang, P., He, Z., Xie, K., Gao, J., & Antolovich, M. (2017). A nonnegative projection based algorithm for low-rank nonnegative matrix approximation. In D. Liu, S. Xie, Y. Li, D. Zhao, & E.-S. M. El-Alfy (Eds.), Neural information processing: 24th International Conference, ICONIP 2017, Guangzhou, China, November 14-18, 2017, Proceedings, Part I (pp. 240-247). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10634 LNCS). Springer-Verlag London Ltd.. https://doi.org/10.1007/978-3-319-70087-8_26

Wang, Peitao ; He, Zhaoshui ; Xie, Kan et al. / A nonnegative projection based algorithm for low-rank nonnegative matrix approximation. Neural information processing: 24th International Conference, ICONIP 2017, Guangzhou, China, November 14-18, 2017, Proceedings, Part I. editor / Derong Liu ; Shengli Xie ; Yuanqing Li ; Dongbin Zhao ; El-Sayed M. El-Alfy. Springer-Verlag London Ltd., 2017. pp. 240-247 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{78abb9af14444b5a9a2bb89eff1b75ae,

title = "A nonnegative projection based algorithm for low-rank nonnegative matrix approximation",

abstract = "Nonnegative matrix factorization/approximation (NMF/NMA) is a widely used method for data analysis. So far, many multiplicative update algorithms have been developed for NMF. In this paper, we propose a nonnegative projection based NMF algorithm, which is different from the conventional multiplicative update NMF algorithms and decreases the objective function by performing Procrustes rotation and nonnegative projection alternately. The experiment results demonstrate that the new algorithm converges much faster than traditional ones.",

keywords = "Alternative updating, Low-rank approximation, Nonnegative matrix factorization",

author = "Peitao Wang and Zhaoshui He and Kan Xie and Junbin Gao and Michael Antolovich",

note = "Includes bibliographical references.; 24th International Conference on Neural Information Processing : ICONIP 2017 ; Conference date: 14-11-2017 Through 18-11-2017",

year = "2017",

month = oct,

day = "24",

doi = "10.1007/978-3-319-70087-8_26",

language = "English",

isbn = "9783319700861",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer-Verlag London Ltd.",

pages = "240--247",

editor = "Derong Liu and Shengli Xie and Yuanqing Li and Dongbin Zhao and El-Alfy, {El-Sayed M.}",

booktitle = "Neural information processing",

address = "Germany",

url = "http://www.conference123.org/iconip2017/index.html, https://link.springer.com/book/10.1007/978-3-319-70087-8#about",

}

Wang, P, He, Z, Xie, K, Gao, J & Antolovich, M 2017, A nonnegative projection based algorithm for low-rank nonnegative matrix approximation. in D Liu, S Xie, Y Li, D Zhao & E-SM El-Alfy (eds), Neural information processing: 24th International Conference, ICONIP 2017, Guangzhou, China, November 14-18, 2017, Proceedings, Part I. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 10634 LNCS, Springer-Verlag London Ltd., pp. 240-247, 24th International Conference on Neural Information Processing, Guangzhou, China, 14/11/17. https://doi.org/10.1007/978-3-319-70087-8_26

A nonnegative projection based algorithm for low-rank nonnegative matrix approximation. / Wang, Peitao; He, Zhaoshui; Xie, Kan et al.
Neural information processing: 24th International Conference, ICONIP 2017, Guangzhou, China, November 14-18, 2017, Proceedings, Part I. ed. / Derong Liu; Shengli Xie; Yuanqing Li; Dongbin Zhao; El-Sayed M. El-Alfy. Springer-Verlag London Ltd., 2017. p. 240-247 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10634 LNCS).

Research output: Book chapter/Published conference paper › Conference paper › peer-review

TY - GEN

T1 - A nonnegative projection based algorithm for low-rank nonnegative matrix approximation

AU - Wang, Peitao

AU - He, Zhaoshui

AU - Xie, Kan

AU - Gao, Junbin

AU - Antolovich, Michael

N1 - Includes bibliographical references.

PY - 2017/10/24

Y1 - 2017/10/24

N2 - Nonnegative matrix factorization/approximation (NMF/NMA) is a widely used method for data analysis. So far, many multiplicative update algorithms have been developed for NMF. In this paper, we propose a nonnegative projection based NMF algorithm, which is different from the conventional multiplicative update NMF algorithms and decreases the objective function by performing Procrustes rotation and nonnegative projection alternately. The experiment results demonstrate that the new algorithm converges much faster than traditional ones.

AB - Nonnegative matrix factorization/approximation (NMF/NMA) is a widely used method for data analysis. So far, many multiplicative update algorithms have been developed for NMF. In this paper, we propose a nonnegative projection based NMF algorithm, which is different from the conventional multiplicative update NMF algorithms and decreases the objective function by performing Procrustes rotation and nonnegative projection alternately. The experiment results demonstrate that the new algorithm converges much faster than traditional ones.

KW - Alternative updating

KW - Low-rank approximation

KW - Nonnegative matrix factorization

UR - http://www.scopus.com/inward/record.url?scp=85035090696&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85035090696&partnerID=8YFLogxK

UR - https://link.springer.com/content/pdf/bfm%3A978-3-319-70087-8%2F1.pdf

UR - http://www.conference123.org/iconip2017/files/ICONIP%202017%20Final%20Program.pdf

U2 - 10.1007/978-3-319-70087-8_26

DO - 10.1007/978-3-319-70087-8_26

M3 - Conference paper

AN - SCOPUS:85035090696

SN - 9783319700861

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 240

EP - 247

BT - Neural information processing

A2 - Liu, Derong

A2 - Xie, Shengli

A2 - Li, Yuanqing

A2 - Zhao, Dongbin

A2 - El-Alfy, El-Sayed M.

PB - Springer-Verlag London Ltd.

T2 - 24th International Conference on Neural Information Processing

Y2 - 14 November 2017 through 18 November 2017

ER -

Wang P, He Z, Xie K, Gao J, Antolovich M. A nonnegative projection based algorithm for low-rank nonnegative matrix approximation. In Liu D, Xie S, Li Y, Zhao D, El-Alfy ESM, editors, Neural information processing: 24th International Conference, ICONIP 2017, Guangzhou, China, November 14-18, 2017, Proceedings, Part I. Springer-Verlag London Ltd. 2017. p. 240-247. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-319-70087-8_26

A nonnegative projection based algorithm for low-rank nonnegative matrix approximation

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this