Feature selection for fatigue segment classification system using elitist non-dominated sorting in genetic algorithm

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

Having relevant features for representing dataset would motivate such algorithms to provide a highly accurate classification system in less-consuming time. Unfortunately, one good set of features is sometimes not fit to all learning algorithms. To confirm that learning algorithm selection does not weights system accuracy user has to validate that the given dataset is a feature-oriented dataset. Thus, in this study we propose a simple verification procedure based on multi objective approach by means of elitist Non-dominated Sorting in Genetic Algorithm (NSGA-II). The way NSGA-II performs in this work is quite similar to the feature selection procedure except on interpretation of the results i.e. set of optimal solutions. Two conflicting minimization elements namely classification error and number of used features are taken as objective functions. A case study of fatigue segment classification was chosen for the purpose of this study where simulations were repeated using four single classifiers such as Naive-Bayes, k nearest neighbours, decision tree and radial basis function. The proposed procedure demonstrates that only two features are needed for classifying a fatigue segment task without having to place concern on learning algorithm.

Original languageEnglish
Title of host publicationApplied Mechanics and Materials
Pages232-236
Number of pages5
Volume165
DOIs
Publication statusPublished - 2012
EventRegional Conference on Automotive Research, ReCAR 2011 - Kuala Lumpur
Duration: 14 Dec 201115 Dec 2011

Publication series

NameApplied Mechanics and Materials
Volume165
ISSN (Print)16609336
ISSN (Electronic)16627482

Other

OtherRegional Conference on Automotive Research, ReCAR 2011
CityKuala Lumpur
Period14/12/1115/12/11

Fingerprint

Sorting
Learning algorithms
Feature extraction
Genetic algorithms
Fatigue of materials
Decision trees
Classifiers

Keywords

  • Classification
  • Fatigue life
  • Feature selection
  • Immune features
  • NSGA-II

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Feature selection for fatigue segment classification system using elitist non-dominated sorting in genetic algorithm. / Osman, Mohd Haniff; Mohd Nopiah, Zulkifli; Abdullah, Shahrum.

Applied Mechanics and Materials. Vol. 165 2012. p. 232-236 (Applied Mechanics and Materials; Vol. 165).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Osman, MH, Mohd Nopiah, Z & Abdullah, S 2012, Feature selection for fatigue segment classification system using elitist non-dominated sorting in genetic algorithm. in Applied Mechanics and Materials. vol. 165, Applied Mechanics and Materials, vol. 165, pp. 232-236, Regional Conference on Automotive Research, ReCAR 2011, Kuala Lumpur, 14/12/11. https://doi.org/10.4028/www.scientific.net/AMM.165.232
@inproceedings{f7578b77a2184ae99e192edcfe0ff753,
title = "Feature selection for fatigue segment classification system using elitist non-dominated sorting in genetic algorithm",
abstract = "Having relevant features for representing dataset would motivate such algorithms to provide a highly accurate classification system in less-consuming time. Unfortunately, one good set of features is sometimes not fit to all learning algorithms. To confirm that learning algorithm selection does not weights system accuracy user has to validate that the given dataset is a feature-oriented dataset. Thus, in this study we propose a simple verification procedure based on multi objective approach by means of elitist Non-dominated Sorting in Genetic Algorithm (NSGA-II). The way NSGA-II performs in this work is quite similar to the feature selection procedure except on interpretation of the results i.e. set of optimal solutions. Two conflicting minimization elements namely classification error and number of used features are taken as objective functions. A case study of fatigue segment classification was chosen for the purpose of this study where simulations were repeated using four single classifiers such as Naive-Bayes, k nearest neighbours, decision tree and radial basis function. The proposed procedure demonstrates that only two features are needed for classifying a fatigue segment task without having to place concern on learning algorithm.",
keywords = "Classification, Fatigue life, Feature selection, Immune features, NSGA-II",
author = "Osman, {Mohd Haniff} and {Mohd Nopiah}, Zulkifli and Shahrum Abdullah",
year = "2012",
doi = "10.4028/www.scientific.net/AMM.165.232",
language = "English",
isbn = "9783037854136",
volume = "165",
series = "Applied Mechanics and Materials",
pages = "232--236",
booktitle = "Applied Mechanics and Materials",

}

TY - GEN

T1 - Feature selection for fatigue segment classification system using elitist non-dominated sorting in genetic algorithm

AU - Osman, Mohd Haniff

AU - Mohd Nopiah, Zulkifli

AU - Abdullah, Shahrum

PY - 2012

Y1 - 2012

N2 - Having relevant features for representing dataset would motivate such algorithms to provide a highly accurate classification system in less-consuming time. Unfortunately, one good set of features is sometimes not fit to all learning algorithms. To confirm that learning algorithm selection does not weights system accuracy user has to validate that the given dataset is a feature-oriented dataset. Thus, in this study we propose a simple verification procedure based on multi objective approach by means of elitist Non-dominated Sorting in Genetic Algorithm (NSGA-II). The way NSGA-II performs in this work is quite similar to the feature selection procedure except on interpretation of the results i.e. set of optimal solutions. Two conflicting minimization elements namely classification error and number of used features are taken as objective functions. A case study of fatigue segment classification was chosen for the purpose of this study where simulations were repeated using four single classifiers such as Naive-Bayes, k nearest neighbours, decision tree and radial basis function. The proposed procedure demonstrates that only two features are needed for classifying a fatigue segment task without having to place concern on learning algorithm.

AB - Having relevant features for representing dataset would motivate such algorithms to provide a highly accurate classification system in less-consuming time. Unfortunately, one good set of features is sometimes not fit to all learning algorithms. To confirm that learning algorithm selection does not weights system accuracy user has to validate that the given dataset is a feature-oriented dataset. Thus, in this study we propose a simple verification procedure based on multi objective approach by means of elitist Non-dominated Sorting in Genetic Algorithm (NSGA-II). The way NSGA-II performs in this work is quite similar to the feature selection procedure except on interpretation of the results i.e. set of optimal solutions. Two conflicting minimization elements namely classification error and number of used features are taken as objective functions. A case study of fatigue segment classification was chosen for the purpose of this study where simulations were repeated using four single classifiers such as Naive-Bayes, k nearest neighbours, decision tree and radial basis function. The proposed procedure demonstrates that only two features are needed for classifying a fatigue segment task without having to place concern on learning algorithm.

KW - Classification

KW - Fatigue life

KW - Feature selection

KW - Immune features

KW - NSGA-II

UR - http://www.scopus.com/inward/record.url?scp=84860792725&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84860792725&partnerID=8YFLogxK

U2 - 10.4028/www.scientific.net/AMM.165.232

DO - 10.4028/www.scientific.net/AMM.165.232

M3 - Conference contribution

SN - 9783037854136

VL - 165

T3 - Applied Mechanics and Materials

SP - 232

EP - 236

BT - Applied Mechanics and Materials

ER -