Support vector machine (SVM) algorithms are a popular class of techniques to perform classification. However, outliers in the data can result in bad global misclassification percentages. In this paper, we propose a method to identify such outliers in the SVM framework. A specific robust classification algorithm is proposed adjusting the least squares SVM (LS-SVM). This yields better classification performance for heavily tailed data and data containing outliers.
%0 Journal Article
%1 verdocnk:2009
%A Debruyne, Michiel
%A Serneels, Sven
%A Verdonck, Tim
%C Department of Mathematics and Computer Science, University of Antwerp, Belgium; LS Services and Consultancy, Edegem, Belgium
%D 2009
%I John Wiley & Sons
%J Journal of Chemometrics
%K SVM classification robust
%N 9
%P 479-486
%R 10.1002/cem.1241
%T Robustified least squares support vector classification
%U http://dx.doi.org/10.1002/cem.1241
%V 23
@article{verdocnk:2009,
added-at = {2009-11-02T23:03:09.000+0100},
address = {Department of Mathematics and Computer Science, University of Antwerp, Belgium; LS Services and Consultancy, Edegem, Belgium},
author = {Debruyne, Michiel and Serneels, Sven and Verdonck, Tim},
biburl = {https://www.bibsonomy.org/bibtex/2357c0cb65f092b70eaa343e7c99d0289/vivion},
description = {Support vector machine (SVM) algorithms are a popular class of techniques to perform classification. However, outliers in the data can result in bad global misclassification percentages. In this paper, we propose a method to identify such outliers in the SVM framework. A specific robust classification algorithm is proposed adjusting the least squares SVM (LS-SVM). This yields better classification performance for heavily tailed data and data containing outliers.},
doi = {10.1002/cem.1241},
interhash = {f4d550cda75cd239e93132e80c2632ef},
intrahash = {357c0cb65f092b70eaa343e7c99d0289},
journal = {Journal of Chemometrics},
keywords = {SVM classification robust},
number = 9,
pages = {479-486},
publisher = {John Wiley & Sons},
timestamp = {2009-11-02T23:03:09.000+0100},
title = {Robustified least squares support vector classification},
url = {http://dx.doi.org/10.1002/cem.1241},
volume = 23,
year = 2009
}