Weighted additive criterion for linear dimension reduction

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

2 Citations (Scopus)

Abstract

Linear discriminant analysis (LDA) for dimension reduction has been applied to a wide variety of face recognition tasks. However, it has two major problems. First, it suffers from the small sample size problem when dimensionality is greater than the sample size. Second, it creates subspaces that favor well separated classes over those that are not. In this paper, we propose a simple weighted criterion for linear dimension reduction that addresses the above two problems associated with LDA. In addition, there are well established numerical procedures such as semi-definite programming for efficiently computing the proposed criterion. We demonstrate the efficacy of our proposal and compare it against other competing techniques using a number of examples.

Original languageEnglish
Title of host publicationProceedings of the 7th IEEE International Conference on Data Mining, ICDM 2007
Pages619-624
Number of pages6
DOIs
StatePublished - 1 Dec 2007
Event7th IEEE International Conference on Data Mining, ICDM 2007 - Omaha, NE, United States
Duration: 28 Oct 200731 Oct 2007

Other

Other7th IEEE International Conference on Data Mining, ICDM 2007
CountryUnited States
CityOmaha, NE
Period28/10/0731/10/07

Fingerprint

Discriminant analysis
Face recognition

Cite this

Peng, J., & Robila, S. (2007). Weighted additive criterion for linear dimension reduction. In Proceedings of the 7th IEEE International Conference on Data Mining, ICDM 2007 (pp. 619-624). [4470300] https://doi.org/10.1109/ICDM.2007.81
Peng, Jing ; Robila, Stefan. / Weighted additive criterion for linear dimension reduction. Proceedings of the 7th IEEE International Conference on Data Mining, ICDM 2007. 2007. pp. 619-624
@inproceedings{ad92b139e3ef4d9288953cf9e844b02b,
title = "Weighted additive criterion for linear dimension reduction",
abstract = "Linear discriminant analysis (LDA) for dimension reduction has been applied to a wide variety of face recognition tasks. However, it has two major problems. First, it suffers from the small sample size problem when dimensionality is greater than the sample size. Second, it creates subspaces that favor well separated classes over those that are not. In this paper, we propose a simple weighted criterion for linear dimension reduction that addresses the above two problems associated with LDA. In addition, there are well established numerical procedures such as semi-definite programming for efficiently computing the proposed criterion. We demonstrate the efficacy of our proposal and compare it against other competing techniques using a number of examples.",
author = "Jing Peng and Stefan Robila",
year = "2007",
month = "12",
day = "1",
doi = "10.1109/ICDM.2007.81",
language = "English",
isbn = "0769530184",
pages = "619--624",
booktitle = "Proceedings of the 7th IEEE International Conference on Data Mining, ICDM 2007",

}

Peng, J & Robila, S 2007, Weighted additive criterion for linear dimension reduction. in Proceedings of the 7th IEEE International Conference on Data Mining, ICDM 2007., 4470300, pp. 619-624, 7th IEEE International Conference on Data Mining, ICDM 2007, Omaha, NE, United States, 28/10/07. https://doi.org/10.1109/ICDM.2007.81

Weighted additive criterion for linear dimension reduction. / Peng, Jing; Robila, Stefan.

Proceedings of the 7th IEEE International Conference on Data Mining, ICDM 2007. 2007. p. 619-624 4470300.

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

TY - GEN

T1 - Weighted additive criterion for linear dimension reduction

AU - Peng, Jing

AU - Robila, Stefan

PY - 2007/12/1

Y1 - 2007/12/1

N2 - Linear discriminant analysis (LDA) for dimension reduction has been applied to a wide variety of face recognition tasks. However, it has two major problems. First, it suffers from the small sample size problem when dimensionality is greater than the sample size. Second, it creates subspaces that favor well separated classes over those that are not. In this paper, we propose a simple weighted criterion for linear dimension reduction that addresses the above two problems associated with LDA. In addition, there are well established numerical procedures such as semi-definite programming for efficiently computing the proposed criterion. We demonstrate the efficacy of our proposal and compare it against other competing techniques using a number of examples.

AB - Linear discriminant analysis (LDA) for dimension reduction has been applied to a wide variety of face recognition tasks. However, it has two major problems. First, it suffers from the small sample size problem when dimensionality is greater than the sample size. Second, it creates subspaces that favor well separated classes over those that are not. In this paper, we propose a simple weighted criterion for linear dimension reduction that addresses the above two problems associated with LDA. In addition, there are well established numerical procedures such as semi-definite programming for efficiently computing the proposed criterion. We demonstrate the efficacy of our proposal and compare it against other competing techniques using a number of examples.

UR - http://www.scopus.com/inward/record.url?scp=49749091148&partnerID=8YFLogxK

U2 - 10.1109/ICDM.2007.81

DO - 10.1109/ICDM.2007.81

M3 - Conference contribution

SN - 0769530184

SN - 9780769530185

SP - 619

EP - 624

BT - Proceedings of the 7th IEEE International Conference on Data Mining, ICDM 2007

ER -

Peng J, Robila S. Weighted additive criterion for linear dimension reduction. In Proceedings of the 7th IEEE International Conference on Data Mining, ICDM 2007. 2007. p. 619-624. 4470300 https://doi.org/10.1109/ICDM.2007.81