Naive Bayes-guided bat algorithm for feature selection

Ahmed Majid Taha, Aida Mustapha, Soong Der Chen

Research output: Contribution to journalArticle

23 Citations (Scopus)

Abstract

When the amount of data and information is said to double in every 20 months or so, feature selection has become highly important and beneficial. Further improvements in feature selection will positively affect a wide array of applications in fields such as pattern recognition, machine learning, or signal processing. Bio-inspired method called Bat Algorithm hybridized with a Naive Bayes classifier has been presented in this work. The performance of the proposed feature selection algorithm was investigated using twelve benchmark datasets from different domains and was compared to three other well-known feature selection algorithms. Discussion focused on four perspectives: number of features, classification accuracy, stability, and feature generalization. The results showed that BANB significantly outperformed other algorithms in selecting lower number of features, hence removing irrelevant, redundant, or noisy features while maintaining the classification accuracy. BANB is also proven to be more stable than other methods and is capable of producing more general feature subsets.

Original languageEnglish
Article number325973
JournalThe Scientific World Journal
Volume2013
DOIs
Publication statusPublished - 01 Dec 2013

Fingerprint

bat
Feature extraction
Benchmarking
signal processing
pattern recognition
Pattern recognition
Learning systems
Signal processing
Classifiers
method

All Science Journal Classification (ASJC) codes

  • Biochemistry, Genetics and Molecular Biology(all)
  • Environmental Science(all)

Cite this

@article{fb36a584a2474c95bb706a824cf6c6cf,
title = "Naive Bayes-guided bat algorithm for feature selection",
abstract = "When the amount of data and information is said to double in every 20 months or so, feature selection has become highly important and beneficial. Further improvements in feature selection will positively affect a wide array of applications in fields such as pattern recognition, machine learning, or signal processing. Bio-inspired method called Bat Algorithm hybridized with a Naive Bayes classifier has been presented in this work. The performance of the proposed feature selection algorithm was investigated using twelve benchmark datasets from different domains and was compared to three other well-known feature selection algorithms. Discussion focused on four perspectives: number of features, classification accuracy, stability, and feature generalization. The results showed that BANB significantly outperformed other algorithms in selecting lower number of features, hence removing irrelevant, redundant, or noisy features while maintaining the classification accuracy. BANB is also proven to be more stable than other methods and is capable of producing more general feature subsets.",
author = "Taha, {Ahmed Majid} and Aida Mustapha and Chen, {Soong Der}",
year = "2013",
month = "12",
day = "1",
doi = "10.1155/2013/325973",
language = "English",
volume = "2013",
journal = "The Scientific World Journal",
issn = "2356-6140",
publisher = "Hindawi Publishing Corporation",

}

Naive Bayes-guided bat algorithm for feature selection. / Taha, Ahmed Majid; Mustapha, Aida; Chen, Soong Der.

In: The Scientific World Journal, Vol. 2013, 325973, 01.12.2013.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Naive Bayes-guided bat algorithm for feature selection

AU - Taha, Ahmed Majid

AU - Mustapha, Aida

AU - Chen, Soong Der

PY - 2013/12/1

Y1 - 2013/12/1

N2 - When the amount of data and information is said to double in every 20 months or so, feature selection has become highly important and beneficial. Further improvements in feature selection will positively affect a wide array of applications in fields such as pattern recognition, machine learning, or signal processing. Bio-inspired method called Bat Algorithm hybridized with a Naive Bayes classifier has been presented in this work. The performance of the proposed feature selection algorithm was investigated using twelve benchmark datasets from different domains and was compared to three other well-known feature selection algorithms. Discussion focused on four perspectives: number of features, classification accuracy, stability, and feature generalization. The results showed that BANB significantly outperformed other algorithms in selecting lower number of features, hence removing irrelevant, redundant, or noisy features while maintaining the classification accuracy. BANB is also proven to be more stable than other methods and is capable of producing more general feature subsets.

AB - When the amount of data and information is said to double in every 20 months or so, feature selection has become highly important and beneficial. Further improvements in feature selection will positively affect a wide array of applications in fields such as pattern recognition, machine learning, or signal processing. Bio-inspired method called Bat Algorithm hybridized with a Naive Bayes classifier has been presented in this work. The performance of the proposed feature selection algorithm was investigated using twelve benchmark datasets from different domains and was compared to three other well-known feature selection algorithms. Discussion focused on four perspectives: number of features, classification accuracy, stability, and feature generalization. The results showed that BANB significantly outperformed other algorithms in selecting lower number of features, hence removing irrelevant, redundant, or noisy features while maintaining the classification accuracy. BANB is also proven to be more stable than other methods and is capable of producing more general feature subsets.

UR - http://www.scopus.com/inward/record.url?scp=84893863229&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84893863229&partnerID=8YFLogxK

U2 - 10.1155/2013/325973

DO - 10.1155/2013/325973

M3 - Article

VL - 2013

JO - The Scientific World Journal

JF - The Scientific World Journal

SN - 2356-6140

M1 - 325973

ER -