Training and analysis of support vector machine using sequential minimal optimization

Training and analysis of support vector machine using sequential minimal optimization

Maximizing the classification performance of the training data is a typical procedure in training a classifier. It is well known that training a Support Vector Machine (SVM) requires the solution of an enormous quadratic programming (QP) optimization problem. Serious challenges appeared in the train...

Full description

Bibliographic Details
Published in:Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics
Main Author: Shahbudin S.; Hussain A.; Samad S.A.; Md. Tahir N.
Format: Conference paper
Language:English
Published: 2008
Online Access:https://www.scopus.com/inward/record.uri?eid=2-s2.0-69949162490&doi=10.1109%2fICSMC.2008.4811304&partnerID=40&md5=ede1f705843f1bdb18195b1b73e8c018
id 2-s2.0-69949162490
spelling 2-s2.0-69949162490
Shahbudin S.; Hussain A.; Samad S.A.; Md. Tahir N.
Training and analysis of support vector machine using sequential minimal optimization
2008
Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics


10.1109/ICSMC.2008.4811304
https://www.scopus.com/inward/record.uri?eid=2-s2.0-69949162490&doi=10.1109%2fICSMC.2008.4811304&partnerID=40&md5=ede1f705843f1bdb18195b1b73e8c018
Maximizing the classification performance of the training data is a typical procedure in training a classifier. It is well known that training a Support Vector Machine (SVM) requires the solution of an enormous quadratic programming (QP) optimization problem. Serious challenges appeared in the training dilemma due to immense training and this could be solved using Sequential Minimal Optimization (SMO). This paper investigates the performance of SMO solver in term of CPU time, number of support vector and decision boundaries when applied in a 2-dimensional datasets. Next, the chunking algorithm is employed for comparison purpose. Initial results demonstrated that the SMO algorithm could enhance the performance of the training dataset. Both algorithms illustrated similar patterns from the decision boundaries attained. Classification rate achieved by both solvers are superb. © 2008 IEEE.

1062922X
English
Conference paper
author Shahbudin S.; Hussain A.; Samad S.A.; Md. Tahir N.
spellingShingle Shahbudin S.; Hussain A.; Samad S.A.; Md. Tahir N.
Training and analysis of support vector machine using sequential minimal optimization
author_facet Shahbudin S.; Hussain A.; Samad S.A.; Md. Tahir N.
author_sort Shahbudin S.; Hussain A.; Samad S.A.; Md. Tahir N.
title Training and analysis of support vector machine using sequential minimal optimization
title_short Training and analysis of support vector machine using sequential minimal optimization
title_full Training and analysis of support vector machine using sequential minimal optimization
title_fullStr Training and analysis of support vector machine using sequential minimal optimization
title_full_unstemmed Training and analysis of support vector machine using sequential minimal optimization
title_sort Training and analysis of support vector machine using sequential minimal optimization
publishDate 2008
container_title Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics
container_volume
container_issue
doi_str_mv 10.1109/ICSMC.2008.4811304
url https://www.scopus.com/inward/record.uri?eid=2-s2.0-69949162490&doi=10.1109%2fICSMC.2008.4811304&partnerID=40&md5=ede1f705843f1bdb18195b1b73e8c018
description Maximizing the classification performance of the training data is a typical procedure in training a classifier. It is well known that training a Support Vector Machine (SVM) requires the solution of an enormous quadratic programming (QP) optimization problem. Serious challenges appeared in the training dilemma due to immense training and this could be solved using Sequential Minimal Optimization (SMO). This paper investigates the performance of SMO solver in term of CPU time, number of support vector and decision boundaries when applied in a 2-dimensional datasets. Next, the chunking algorithm is employed for comparison purpose. Initial results demonstrated that the SMO algorithm could enhance the performance of the training dataset. Both algorithms illustrated similar patterns from the decision boundaries attained. Classification rate achieved by both solvers are superb. © 2008 IEEE.
publisher
issn 1062922X
language English
format Conference paper
accesstype
record_format scopus
collection Scopus
_version_ 1825722586554695680

Similar Items