Quasi-Linear Support Vector Machine for Nonlinear Classification
Summary :
This paper proposes a so called quasi-linear support vector machine (SVM), which is an SVM with a composite quasi-linear kernel. In the quasi-linear SVM model, the nonlinear separation hyperplane is approximated by multiple local linear models with interpolation. Instead of building multiple local SVM models separately, the quasi-linear SVM realizes the multi local linear model approach in the kernel level. That is, it is built exactly in the same way as a single SVM model, by composing a quasi-linear kernel. A guided partitioning method is proposed to obtain the local partitions for the composition of quasi-linear kernel function. Experiment results on artificial data and benchmark datasets show that the proposed method is effective and improves classification performances.
- Publication
- IEICE TRANSACTIONS on Fundamentals Vol.E97-A No.7 pp.1587-1594
- Publication Date
- 2014/07/01
- Publicized
- Online ISSN
- 1745-1337
- DOI
- 10.1587/transfun.E97.A.1587
- Type of Manuscript
- PAPER
- Category
- Neural Networks and Bioengineering
Authors
Bo ZHOU
Waseda University
Benhui CHEN
Dali University
Jinglu HU
Waseda University
Keyword
Latest Issue
Copyrights notice of machine-translated contents
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Cite this
Copy
Bo ZHOU, Benhui CHEN, Jinglu HU, "Quasi-Linear Support Vector Machine for Nonlinear Classification" in IEICE TRANSACTIONS on Fundamentals,
vol. E97-A, no. 7, pp. 1587-1594, July 2014, doi: 10.1587/transfun.E97.A.1587.
Abstract: This paper proposes a so called quasi-linear support vector machine (SVM), which is an SVM with a composite quasi-linear kernel. In the quasi-linear SVM model, the nonlinear separation hyperplane is approximated by multiple local linear models with interpolation. Instead of building multiple local SVM models separately, the quasi-linear SVM realizes the multi local linear model approach in the kernel level. That is, it is built exactly in the same way as a single SVM model, by composing a quasi-linear kernel. A guided partitioning method is proposed to obtain the local partitions for the composition of quasi-linear kernel function. Experiment results on artificial data and benchmark datasets show that the proposed method is effective and improves classification performances.
URL: https://global.ieice.org/en_transactions/fundamentals/10.1587/transfun.E97.A.1587/_p
Copy
@ARTICLE{e97-a_7_1587,
author={Bo ZHOU, Benhui CHEN, Jinglu HU, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Quasi-Linear Support Vector Machine for Nonlinear Classification},
year={2014},
volume={E97-A},
number={7},
pages={1587-1594},
abstract={This paper proposes a so called quasi-linear support vector machine (SVM), which is an SVM with a composite quasi-linear kernel. In the quasi-linear SVM model, the nonlinear separation hyperplane is approximated by multiple local linear models with interpolation. Instead of building multiple local SVM models separately, the quasi-linear SVM realizes the multi local linear model approach in the kernel level. That is, it is built exactly in the same way as a single SVM model, by composing a quasi-linear kernel. A guided partitioning method is proposed to obtain the local partitions for the composition of quasi-linear kernel function. Experiment results on artificial data and benchmark datasets show that the proposed method is effective and improves classification performances.},
keywords={},
doi={10.1587/transfun.E97.A.1587},
ISSN={1745-1337},
month={July},}
Copy
TY - JOUR
TI - Quasi-Linear Support Vector Machine for Nonlinear Classification
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 1587
EP - 1594
AU - Bo ZHOU
AU - Benhui CHEN
AU - Jinglu HU
PY - 2014
DO - 10.1587/transfun.E97.A.1587
JO - IEICE TRANSACTIONS on Fundamentals
SN - 1745-1337
VL - E97-A
IS - 7
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - July 2014
AB - This paper proposes a so called quasi-linear support vector machine (SVM), which is an SVM with a composite quasi-linear kernel. In the quasi-linear SVM model, the nonlinear separation hyperplane is approximated by multiple local linear models with interpolation. Instead of building multiple local SVM models separately, the quasi-linear SVM realizes the multi local linear model approach in the kernel level. That is, it is built exactly in the same way as a single SVM model, by composing a quasi-linear kernel. A guided partitioning method is proposed to obtain the local partitions for the composition of quasi-linear kernel function. Experiment results on artificial data and benchmark datasets show that the proposed method is effective and improves classification performances.
ER -
English
