Bezier Function Smooth Support Vector Machine for Classification

被引：0

作者：

Fan, X. H. ^{[1
]}

Zhang, J. ^{[1
]}

Ma, H. B. ^{[1
]}

机构：

[1] Northwestern Polytech Univ, Sch Elect & Informat, Xian, Peoples R China

来源：

INTERNATIONAL CONFERENCE ON ADVANCED MANAGEMENT SCIENCE AND INFORMATION ENGINEERING (AMSIE 2015) | 2015年

关键词：

pattern recognition; support vector machine; smooth method; Bezier function; classification;

D O I：

暂无

中图分类号：

TP39 [计算机的应用];

学科分类号：

081203 ; 0835 ;

摘要：

In this paper, a new smooth method that Bezier function is used to smoothen the model of support vector machine (SVM) is presented. A class of 1-norm Bezier function support vector machine (BSSVM1) is obtained. Newton-Armijo algorithm is applied to solve the BSSVM1 model. Moreover, our theoretical analysis and numerical experiments confirm that the BSSVM1 model have a better classification performance than other smooth models.

引用

页码：678 / 684

页数：7

共 8 条

[1] [Anonymous], NATURE STATISTI810
[2] Hardy Godfrey Harold, 2009, COURSE PURE MATH
[3] SSVM: A smooth support vector machine for classification
Lee, YJ
Mangasarian, OL
[J]. COMPUTATIONAL OPTIMIZATION AND APPLICATIONS, 2001, 20 (01) : 5 - 22
[4] Li Guang-ming, 2009, Journal of Computer Applications, V29, P1612, DOI 10.3724/SP.J.1087.2009.01612
[5] Musicant D. R., NDC: Normally distributed clustered datasets
[6] Yuan YB, 2005, LECT NOTES ARTIF INT, V3584, P157
[7] Yuan Yu-Bo, 2005, Chinese Journal of Computers, V28, P9
[8] Yuan YB, 2007, J IND MANAG OPTIM, V3, P529

← 1 →