Staff View: A Hybrid Gini PSO-SVM Feature Selection: An Empirical Study of Population Sizes on Different Classifier

A Hybrid Gini PSO-SVM Feature Selection: An Empirical Study of Population Sizes on Different Classifier

A performance of anti-spam filter not only depends on the number of features and types of classifier that are used, but it also depends on the other parameter settings. Deriving from previous experiments, we extended our work by investigating the effect of population sizes from our proposed met...

Full description

Saved in:

Bibliographic Details
Main Authors:	Noormadinah Allias, Megat NorulAzmi Megat Mohamed Noor, Mohd. Nazri Ismail, Kim de Silva, (UniKL MIIT)
Format:
Published:	2014
Subjects:	swarm size taguchi method orthogonal array learning algorithms
Online Access:	http://localhost/xmlui/handle/123456789/6305
Tags:	Add Tag No Tags, Be the first to tag this record!

id	my.unikl.ir-6305
record_format	eprints
spelling	my.unikl.ir-63052014-04-24T02:56:16Z A Hybrid Gini PSO-SVM Feature Selection: An Empirical Study of Population Sizes on Different Classifier Noormadinah Allias Megat NorulAzmi Megat Mohamed Noor Mohd. Nazri Ismail Kim de Silva (UniKL MIIT) swarm size taguchi method orthogonal array learning algorithms A performance of anti-spam filter not only depends on the number of features and types of classifier that are used, but it also depends on the other parameter settings. Deriving from previous experiments, we extended our work by investigating the effect of population sizes from our proposed method of feature selection on different learning classifier algorithms using Random Forest, Voting, Decision Tree, Support Vector Machine and Stacking. The experiment was conducted on Ling-Spam email dataset. The results showed that the Decision Tree with the smallest size of population is able to give the best result compared to NB, SVM, RF, stacking and voting.A performance of anti-spam filter not only depends on the number of features and types of classifier that are used, but it also depends on the other parameter settings. Deriving from previous experiments, we extended our work by investigating the effect of population sizes from our proposed method of feature selection on different learning classifier algorithms using Random Forest, Voting, Decision Tree, Support Vector Machine and Stacking. The experiment was conducted on Ling-Spam email dataset. The results showed that the Decision Tree with the smallest size of population is able to give the best result compared to NB, SVM, RF, stacking and voting. 2014-04-24T02:56:16Z 2014-04-24T02:56:16Z 2014-04-24 http://localhost/xmlui/handle/123456789/6305 Proceeding of: International Conference on Artificial Intelligence, Modelling & Simulation;
institution	Universiti Kuala Lumpur
building	UniKL Library
collection	Institutional Repository
continent	Asia
country	Malaysia
content_provider	Universiti Kuala Lumpur
content_source	UniKL Institutional Repository
url_provider	http://ir.unikl.edu.my/
topic	swarm size taguchi method orthogonal array learning algorithms
spellingShingle	swarm size taguchi method orthogonal array learning algorithms Noormadinah Allias Megat NorulAzmi Megat Mohamed Noor Mohd. Nazri Ismail Kim de Silva (UniKL MIIT) A Hybrid Gini PSO-SVM Feature Selection: An Empirical Study of Population Sizes on Different Classifier
description	A performance of anti-spam filter not only depends on the number of features and types of classifier that are used, but it also depends on the other parameter settings. Deriving from previous experiments, we extended our work by investigating the effect of population sizes from our proposed method of feature selection on different learning classifier algorithms using Random Forest, Voting, Decision Tree, Support Vector Machine and Stacking. The experiment was conducted on Ling-Spam email dataset. The results showed that the Decision Tree with the smallest size of population is able to give the best result compared to NB, SVM, RF, stacking and voting.A performance of anti-spam filter not only depends on the number of features and types of classifier that are used, but it also depends on the other parameter settings. Deriving from previous experiments, we extended our work by investigating the effect of population sizes from our proposed method of feature selection on different learning classifier algorithms using Random Forest, Voting, Decision Tree, Support Vector Machine and Stacking. The experiment was conducted on Ling-Spam email dataset. The results showed that the Decision Tree with the smallest size of population is able to give the best result compared to NB, SVM, RF, stacking and voting.
format
author	Noormadinah Allias Megat NorulAzmi Megat Mohamed Noor Mohd. Nazri Ismail Kim de Silva (UniKL MIIT)
author_facet	Noormadinah Allias Megat NorulAzmi Megat Mohamed Noor Mohd. Nazri Ismail Kim de Silva (UniKL MIIT)
author_sort	Noormadinah Allias
title	A Hybrid Gini PSO-SVM Feature Selection: An Empirical Study of Population Sizes on Different Classifier
title_short	A Hybrid Gini PSO-SVM Feature Selection: An Empirical Study of Population Sizes on Different Classifier
title_full	A Hybrid Gini PSO-SVM Feature Selection: An Empirical Study of Population Sizes on Different Classifier
title_fullStr	A Hybrid Gini PSO-SVM Feature Selection: An Empirical Study of Population Sizes on Different Classifier
title_full_unstemmed	A Hybrid Gini PSO-SVM Feature Selection: An Empirical Study of Population Sizes on Different Classifier
title_sort	hybrid gini pso-svm feature selection: an empirical study of population sizes on different classifier
publishDate	2014
url	http://localhost/xmlui/handle/123456789/6305
_version_	1644484807856488448
score	13.222552

A Hybrid Gini PSO-SVM Feature Selection: An Empirical Study of Population Sizes on Different Classifier

Similar Items