Quintic spline smooth semi-supervised support vector classification machine

Xiaodan Zhang, Jinggai Ma, Aihua Li, Ang Li

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

A semi-supervised vector machine is a relatively new learning method using both labeled and unlabeled data in classification. Since the objective function of the model for an unstrained semi-supervised vector machine is not smooth, many fast optimization algorithms cannot be applied to solve the model. In order to overcome the difficulty of dealing with non-smooth objective functions, new methods that can solve the semi-supervised vector machine with desired classification accuracy are in great demand. A quintic spline function with three-times differentiability at the origin is constructed by a general three-moment method, which can be used to approximate the symmetric hinge loss function. The approximate accuracy of the quintic spline function is estimated. Moreover, a quintic spline smooth semi-support vector machine is obtained and the convergence accuracy of the smooth model to the non-smooth one is analyzed. Three experiments are performed to test the efficiency of the model. The experimental results show that the new model outperforms other smooth models, in terms of classification performance. Furthermore, the new model is not sensitive to the increasing number of the labeled samples, which means that the new model is more efficient.

Original languageEnglish
Article number7170021
Pages (from-to)626-632
Number of pages7
JournalJournal of Systems Engineering and Electronics
Volume26
Issue number3
DOIs
StatePublished - 1 Jun 2015

Fingerprint

Splines
Hinges
Method of moments
Support vector machines

Keywords

  • Convergence
  • quintic spline function
  • semi-supervised
  • smooth
  • support vector classification machine

Cite this

@article{fcd5ea1e4e474e0284e2b44d433fc28b,
title = "Quintic spline smooth semi-supervised support vector classification machine",
abstract = "A semi-supervised vector machine is a relatively new learning method using both labeled and unlabeled data in classification. Since the objective function of the model for an unstrained semi-supervised vector machine is not smooth, many fast optimization algorithms cannot be applied to solve the model. In order to overcome the difficulty of dealing with non-smooth objective functions, new methods that can solve the semi-supervised vector machine with desired classification accuracy are in great demand. A quintic spline function with three-times differentiability at the origin is constructed by a general three-moment method, which can be used to approximate the symmetric hinge loss function. The approximate accuracy of the quintic spline function is estimated. Moreover, a quintic spline smooth semi-support vector machine is obtained and the convergence accuracy of the smooth model to the non-smooth one is analyzed. Three experiments are performed to test the efficiency of the model. The experimental results show that the new model outperforms other smooth models, in terms of classification performance. Furthermore, the new model is not sensitive to the increasing number of the labeled samples, which means that the new model is more efficient.",
keywords = "Convergence, quintic spline function, semi-supervised, smooth, support vector classification machine",
author = "Xiaodan Zhang and Jinggai Ma and Aihua Li and Ang Li",
year = "2015",
month = "6",
day = "1",
doi = "10.1109/JSEE.2015.00070",
language = "English",
volume = "26",
pages = "626--632",
journal = "Journal of Systems Engineering and Electronics",
issn = "1004-4132",
number = "3",

}

Quintic spline smooth semi-supervised support vector classification machine. / Zhang, Xiaodan; Ma, Jinggai; Li, Aihua; Li, Ang.

In: Journal of Systems Engineering and Electronics, Vol. 26, No. 3, 7170021, 01.06.2015, p. 626-632.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Quintic spline smooth semi-supervised support vector classification machine

AU - Zhang, Xiaodan

AU - Ma, Jinggai

AU - Li, Aihua

AU - Li, Ang

PY - 2015/6/1

Y1 - 2015/6/1

N2 - A semi-supervised vector machine is a relatively new learning method using both labeled and unlabeled data in classification. Since the objective function of the model for an unstrained semi-supervised vector machine is not smooth, many fast optimization algorithms cannot be applied to solve the model. In order to overcome the difficulty of dealing with non-smooth objective functions, new methods that can solve the semi-supervised vector machine with desired classification accuracy are in great demand. A quintic spline function with three-times differentiability at the origin is constructed by a general three-moment method, which can be used to approximate the symmetric hinge loss function. The approximate accuracy of the quintic spline function is estimated. Moreover, a quintic spline smooth semi-support vector machine is obtained and the convergence accuracy of the smooth model to the non-smooth one is analyzed. Three experiments are performed to test the efficiency of the model. The experimental results show that the new model outperforms other smooth models, in terms of classification performance. Furthermore, the new model is not sensitive to the increasing number of the labeled samples, which means that the new model is more efficient.

AB - A semi-supervised vector machine is a relatively new learning method using both labeled and unlabeled data in classification. Since the objective function of the model for an unstrained semi-supervised vector machine is not smooth, many fast optimization algorithms cannot be applied to solve the model. In order to overcome the difficulty of dealing with non-smooth objective functions, new methods that can solve the semi-supervised vector machine with desired classification accuracy are in great demand. A quintic spline function with three-times differentiability at the origin is constructed by a general three-moment method, which can be used to approximate the symmetric hinge loss function. The approximate accuracy of the quintic spline function is estimated. Moreover, a quintic spline smooth semi-support vector machine is obtained and the convergence accuracy of the smooth model to the non-smooth one is analyzed. Three experiments are performed to test the efficiency of the model. The experimental results show that the new model outperforms other smooth models, in terms of classification performance. Furthermore, the new model is not sensitive to the increasing number of the labeled samples, which means that the new model is more efficient.

KW - Convergence

KW - quintic spline function

KW - semi-supervised

KW - smooth

KW - support vector classification machine

UR - http://www.scopus.com/inward/record.url?scp=84938576858&partnerID=8YFLogxK

U2 - 10.1109/JSEE.2015.00070

DO - 10.1109/JSEE.2015.00070

M3 - Article

AN - SCOPUS:84938576858

VL - 26

SP - 626

EP - 632

JO - Journal of Systems Engineering and Electronics

JF - Journal of Systems Engineering and Electronics

SN - 1004-4132

IS - 3

M1 - 7170021

ER -