An improvement on parametric -support vector algorithm for classification

One effective technique that has recently been considered for solving classification problems is parametric [Formula omitted]-support vector regression. This method obtains a concurrent learning framework for both margin determination and function approximation and leads to a convex quadratic progra...

Full description

Saved in:
Bibliographic Details
Published in:Annals of operations research 2019-05, Vol.276 (1-2), p.155-168
Main Authors: Ketabchi, Saeed, Moosaei, Hossein, Razzaghi, Mohamad, Pardalos, Panos M
Format: Article
Language:English
Subjects:
Analysis
Classification
Operations research
Optimization theory
Parametric equations
Quadratic programming
Studies
Support vector machines
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:One effective technique that has recently been considered for solving classification problems is parametric [Formula omitted]-support vector regression. This method obtains a concurrent learning framework for both margin determination and function approximation and leads to a convex quadratic programming problem. In this paper we introduce a new idea that converts this problem into an unconstrained convex problem. Moreover, we propose an extension of Newton's method for solving the unconstrained convex problem. We compare the accuracy and efficiency of our method with support vector machines and parametric [Formula omitted]-support vector regression methods. Experimental results on several UCI benchmark data sets indicate the high efficiency and accuracy of this method.
ISSN:0254-5330
1572-9338
DOI:10.1007/s10479-017-2724-8