Towards evolutionary nonnegative matrix factorization

Fei Wang, Hanghang Tong, Ching Yung Lin

Research output: Chapter in Book/Report/Conference proceedingConference contribution

9 Citations (Scopus)

Abstract

Nonnegative Matrix Factorization (NMF) techniques has aroused considerable interests from the field of artificial intelligence in recent years because of its good interpretability and computational efficiency. However, in many real world applications, the data features usually evolve over time smoothly. In this case, it would be very expensive in both computation and storage to rerun the whole NMF procedure after each time when the data feature changing. In this paper, we propose Evolutionary Nonnegative Matrix Factorization (eNMF), which aims to incrementally update the factorized matrices in a computation and space efficient manner with the variation of the data matrix. We devise such evolutionary procedure for both asymmetric and symmetric NMF. Finally we conduct experiments on several real world data sets to demonstrate the efficacy and efficiency of eNMF.

Original languageEnglish (US)
Title of host publicationProceedings of the National Conference on Artificial Intelligence
Pages501-506
Number of pages6
Volume1
StatePublished - 2011
Externally publishedYes
Event25th AAAI Conference on Artificial Intelligence and the 23rd Innovative Applications of Artificial Intelligence Conference, AAAI-11 / IAAI-11 - San Francisco, CA, United States
Duration: Aug 7 2011Aug 11 2011

Other

Other25th AAAI Conference on Artificial Intelligence and the 23rd Innovative Applications of Artificial Intelligence Conference, AAAI-11 / IAAI-11
CountryUnited States
CitySan Francisco, CA
Period8/7/118/11/11

Fingerprint

Factorization
Computational efficiency
Artificial intelligence
Experiments

ASJC Scopus subject areas

  • Software
  • Artificial Intelligence

Cite this

Wang, F., Tong, H., & Lin, C. Y. (2011). Towards evolutionary nonnegative matrix factorization. In Proceedings of the National Conference on Artificial Intelligence (Vol. 1, pp. 501-506)

Towards evolutionary nonnegative matrix factorization. / Wang, Fei; Tong, Hanghang; Lin, Ching Yung.

Proceedings of the National Conference on Artificial Intelligence. Vol. 1 2011. p. 501-506.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Wang, F, Tong, H & Lin, CY 2011, Towards evolutionary nonnegative matrix factorization. in Proceedings of the National Conference on Artificial Intelligence. vol. 1, pp. 501-506, 25th AAAI Conference on Artificial Intelligence and the 23rd Innovative Applications of Artificial Intelligence Conference, AAAI-11 / IAAI-11, San Francisco, CA, United States, 8/7/11.
Wang F, Tong H, Lin CY. Towards evolutionary nonnegative matrix factorization. In Proceedings of the National Conference on Artificial Intelligence. Vol. 1. 2011. p. 501-506
Wang, Fei ; Tong, Hanghang ; Lin, Ching Yung. / Towards evolutionary nonnegative matrix factorization. Proceedings of the National Conference on Artificial Intelligence. Vol. 1 2011. pp. 501-506
@inproceedings{95d6edbd6c8d431caea0db72a0288381,
title = "Towards evolutionary nonnegative matrix factorization",
abstract = "Nonnegative Matrix Factorization (NMF) techniques has aroused considerable interests from the field of artificial intelligence in recent years because of its good interpretability and computational efficiency. However, in many real world applications, the data features usually evolve over time smoothly. In this case, it would be very expensive in both computation and storage to rerun the whole NMF procedure after each time when the data feature changing. In this paper, we propose Evolutionary Nonnegative Matrix Factorization (eNMF), which aims to incrementally update the factorized matrices in a computation and space efficient manner with the variation of the data matrix. We devise such evolutionary procedure for both asymmetric and symmetric NMF. Finally we conduct experiments on several real world data sets to demonstrate the efficacy and efficiency of eNMF.",
author = "Fei Wang and Hanghang Tong and Lin, {Ching Yung}",
year = "2011",
language = "English (US)",
isbn = "9781577355083",
volume = "1",
pages = "501--506",
booktitle = "Proceedings of the National Conference on Artificial Intelligence",

}

TY - GEN

T1 - Towards evolutionary nonnegative matrix factorization

AU - Wang, Fei

AU - Tong, Hanghang

AU - Lin, Ching Yung

PY - 2011

Y1 - 2011

N2 - Nonnegative Matrix Factorization (NMF) techniques has aroused considerable interests from the field of artificial intelligence in recent years because of its good interpretability and computational efficiency. However, in many real world applications, the data features usually evolve over time smoothly. In this case, it would be very expensive in both computation and storage to rerun the whole NMF procedure after each time when the data feature changing. In this paper, we propose Evolutionary Nonnegative Matrix Factorization (eNMF), which aims to incrementally update the factorized matrices in a computation and space efficient manner with the variation of the data matrix. We devise such evolutionary procedure for both asymmetric and symmetric NMF. Finally we conduct experiments on several real world data sets to demonstrate the efficacy and efficiency of eNMF.

AB - Nonnegative Matrix Factorization (NMF) techniques has aroused considerable interests from the field of artificial intelligence in recent years because of its good interpretability and computational efficiency. However, in many real world applications, the data features usually evolve over time smoothly. In this case, it would be very expensive in both computation and storage to rerun the whole NMF procedure after each time when the data feature changing. In this paper, we propose Evolutionary Nonnegative Matrix Factorization (eNMF), which aims to incrementally update the factorized matrices in a computation and space efficient manner with the variation of the data matrix. We devise such evolutionary procedure for both asymmetric and symmetric NMF. Finally we conduct experiments on several real world data sets to demonstrate the efficacy and efficiency of eNMF.

UR - http://www.scopus.com/inward/record.url?scp=80055037014&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=80055037014&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:80055037014

SN - 9781577355083

VL - 1

SP - 501

EP - 506

BT - Proceedings of the National Conference on Artificial Intelligence

ER -