On L1-norm multi-class support vector machines

Lifeng Wang; Xiaotong T Shen; Yuan F. Zheng

doi:10.1109/ICMLA.2006.38

On L₁-norm multi-class support vector machines

Lifeng Wang
, Xiaotong T Shen
, Yuan F. Zheng

Statistics (Twin Cities)

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

5 Scopus citations

Abstract

Binary Support Vector Machines (SVM) have proven effective in classification. However, problems remain with respect to feature selection in multi-class classification. This article proposes a novel multi-class SVM, which performs classification and feature selection simultaneously via Li-norm penalized sparse representations. The proposed methodology, together with our developed regularization solution path, permits feature selection within the framework of classification. The operational characteristics of the proposed methodology is examined via both simulated and benchmark exampies, and is compared to some competitors in terms of the accuracy of prediction and feature selection. The numerical results suggest that the proposed methodology is highly competitive.

Original language	English (US)
Title of host publication	Proceedings - 5th International Conference on Machine Learning and Applications, ICMLA 2006
Pages	83-88
Number of pages	6
DOIs	https://doi.org/10.1109/ICMLA.2006.38
State	Published - 2006
Event	5th International Conference on Machine Learning and Applications, ICMLA 2006 - Orlando, FL, United States Duration: Dec 14 2006 → Dec 16 2006

Publication series

Name	Proceedings - 5th International Conference on Machine Learning and Applications, ICMLA 2006

Other

Other	5th International Conference on Machine Learning and Applications, ICMLA 2006
Country/Territory	United States
City	Orlando, FL
Period	12/14/06 → 12/16/06

Publisher link

10.1109/ICMLA.2006.38

Find It

Find It at U of M Libraries

Cite this

On L₁-norm multi-class support vector machines. / Wang, Lifeng; Shen, Xiaotong T; Zheng, Yuan F.
Proceedings - 5th International Conference on Machine Learning and Applications, ICMLA 2006. 2006. p. 83-88 4041474 (Proceedings - 5th International Conference on Machine Learning and Applications, ICMLA 2006).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Wang, L, Shen, XT & Zheng, YF 2006, On L₁-norm multi-class support vector machines. in Proceedings - 5th International Conference on Machine Learning and Applications, ICMLA 2006., 4041474, Proceedings - 5th International Conference on Machine Learning and Applications, ICMLA 2006, pp. 83-88, 5th International Conference on Machine Learning and Applications, ICMLA 2006, Orlando, FL, United States, 12/14/06. https://doi.org/10.1109/ICMLA.2006.38

@inproceedings{e0e9fcdd0c2b48d28ac0129e0ffa94ce,

title = "On L1-norm multi-class support vector machines",

abstract = "Binary Support Vector Machines (SVM) have proven effective in classification. However, problems remain with respect to feature selection in multi-class classification. This article proposes a novel multi-class SVM, which performs classification and feature selection simultaneously via Li-norm penalized sparse representations. The proposed methodology, together with our developed regularization solution path, permits feature selection within the framework of classification. The operational characteristics of the proposed methodology is examined via both simulated and benchmark exampies, and is compared to some competitors in terms of the accuracy of prediction and feature selection. The numerical results suggest that the proposed methodology is highly competitive.",

author = "Lifeng Wang and Shen, \{Xiaotong T\} and Zheng, \{Yuan F.\}",

year = "2006",

doi = "10.1109/ICMLA.2006.38",

language = "English (US)",

isbn = "0769527353",

series = "Proceedings - 5th International Conference on Machine Learning and Applications, ICMLA 2006",

pages = "83--88",

booktitle = "Proceedings - 5th International Conference on Machine Learning and Applications, ICMLA 2006",

note = "5th International Conference on Machine Learning and Applications, ICMLA 2006 ; Conference date: 14-12-2006 Through 16-12-2006",

}

TY - GEN

T1 - On L1-norm multi-class support vector machines

AU - Wang, Lifeng

AU - Shen, Xiaotong T

AU - Zheng, Yuan F.

PY - 2006

Y1 - 2006

N2 - Binary Support Vector Machines (SVM) have proven effective in classification. However, problems remain with respect to feature selection in multi-class classification. This article proposes a novel multi-class SVM, which performs classification and feature selection simultaneously via Li-norm penalized sparse representations. The proposed methodology, together with our developed regularization solution path, permits feature selection within the framework of classification. The operational characteristics of the proposed methodology is examined via both simulated and benchmark exampies, and is compared to some competitors in terms of the accuracy of prediction and feature selection. The numerical results suggest that the proposed methodology is highly competitive.

AB - Binary Support Vector Machines (SVM) have proven effective in classification. However, problems remain with respect to feature selection in multi-class classification. This article proposes a novel multi-class SVM, which performs classification and feature selection simultaneously via Li-norm penalized sparse representations. The proposed methodology, together with our developed regularization solution path, permits feature selection within the framework of classification. The operational characteristics of the proposed methodology is examined via both simulated and benchmark exampies, and is compared to some competitors in terms of the accuracy of prediction and feature selection. The numerical results suggest that the proposed methodology is highly competitive.

UR - https://www.scopus.com/pages/publications/40349093379

UR - https://www.scopus.com/pages/publications/40349093379#tab=citedBy

U2 - 10.1109/ICMLA.2006.38

DO - 10.1109/ICMLA.2006.38

M3 - Conference contribution

AN - SCOPUS:40349093379

SN - 0769527353

SN - 9780769527352

T3 - Proceedings - 5th International Conference on Machine Learning and Applications, ICMLA 2006

SP - 83

EP - 88

BT - Proceedings - 5th International Conference on Machine Learning and Applications, ICMLA 2006

T2 - 5th International Conference on Machine Learning and Applications, ICMLA 2006

Y2 - 14 December 2006 through 16 December 2006

ER -