Alternative Strategies in Learning Nonlinear Soft Margin Support Vector Machines

The aims of the paper are multifold, to propose a new method to determine a suitable value of the bias corresponding to the soft margin SVM classifier and to experimentally evaluate the quality of the found value against one of the standard expression of the bias computed in terms of the support vec...

Full description

Saved in:
Bibliographic Details
Published inInformatica Economica Vol. 18; no. 2; pp. 42 - 52
Main Authors Cocianu, Catalina, State, Luminita, Uscatu, Cristian
Format Journal Article
LanguageEnglish
Published Bucharest INFOREC Association 01.01.2014
Inforec Association
Subjects
Algebra
Algorithms
Bias
Classifier Design and Evaluation
Comparative analysis
Computation
Economics
Genetic Search
Kernel Function
Learning
Mathematical analysis
Non-Linear Support Vector Machines
Nonlinear equations
Radial Basis Function
Searching
SMO Platt’s Algorithm
Soft Margin SVM
Studies
Support vector machines
Online AccessGet full text

Cover

More Information
Summary:The aims of the paper are multifold, to propose a new method to determine a suitable value of the bias corresponding to the soft margin SVM classifier and to experimentally evaluate the quality of the found value against one of the standard expression of the bias computed in terms of the support vectors. Also, it is proposed a variant of the Platt's SMO algorithm to compute an approximation of the optimal solution of the SVM QP-problem. The new method for computing a more suitable value of the bias is based on genetic search. In order to evaluate the quality of the proposed method from the point of view of recognition and generalization rates, several tests were performed, some of the results being reported in the final section of the paper.
Bibliography:ObjectType-Article-2
SourceType-Scholarly Journals-1
ObjectType-Feature-1
content type line 23
ISSN:1453-1305
1842-8088
DOI:10.12948/issn14531305/18.2.2014.05