A maximum-margin genetic algorithm for misclassification cost minimizing feature selection problem

Research output: Contribution to journalArticle

9 Citations (Scopus)

Abstract

We consider a feature selection problem where the decision-making objective is to minimize overall misclassification cost by selecting relevant features from a training dataset. We propose a two-stage solution approach for solving misclassification cost minimizing feature selection (MCMFS) problem. Additionally, we propose a maximum-margin genetic algorithm (MMGA) that maximizes margin of separation between classes by taking into account all examples as opposed to maximizing margin of separation using a few support vectors. Feature selection is carried out by either an exhaustive or a heuristic simulated annealing approach in the first stage and a cost sensitive classification using either MMGA or cost sensitive support vector machines (SVM) in the second stage. Using simulated and real-world data sets and different misclassification cost matrices, we test our two-stage approach for solving the MCMFS problem. Our results indicate that feature selection plays an important role when misclassification cost asymmetries increase and the MMGA shows equal or better performance than the SVM.

Original languageEnglish (US)
Pages (from-to)3918-3925
Number of pages8
JournalExpert Systems with Applications
Volume40
Issue number10
DOIs
StatePublished - Aug 1 2013

Fingerprint

Feature extraction
Genetic algorithms
Costs
Support vector machines
Simulated annealing
Decision making

All Science Journal Classification (ASJC) codes

  • Engineering(all)
  • Computer Science Applications
  • Artificial Intelligence

Cite this

@article{10da82bdacf8496cae3f13eaf56eb9d4,
title = "A maximum-margin genetic algorithm for misclassification cost minimizing feature selection problem",
abstract = "We consider a feature selection problem where the decision-making objective is to minimize overall misclassification cost by selecting relevant features from a training dataset. We propose a two-stage solution approach for solving misclassification cost minimizing feature selection (MCMFS) problem. Additionally, we propose a maximum-margin genetic algorithm (MMGA) that maximizes margin of separation between classes by taking into account all examples as opposed to maximizing margin of separation using a few support vectors. Feature selection is carried out by either an exhaustive or a heuristic simulated annealing approach in the first stage and a cost sensitive classification using either MMGA or cost sensitive support vector machines (SVM) in the second stage. Using simulated and real-world data sets and different misclassification cost matrices, we test our two-stage approach for solving the MCMFS problem. Our results indicate that feature selection plays an important role when misclassification cost asymmetries increase and the MMGA shows equal or better performance than the SVM.",
author = "Pendharkar, {Parag C.}",
year = "2013",
month = "8",
day = "1",
doi = "10.1016/j.eswa.2012.12.086",
language = "English (US)",
volume = "40",
pages = "3918--3925",
journal = "Expert Systems with Applications",
issn = "0957-4174",
publisher = "Elsevier Limited",
number = "10",

}

A maximum-margin genetic algorithm for misclassification cost minimizing feature selection problem. / Pendharkar, Parag C.

In: Expert Systems with Applications, Vol. 40, No. 10, 01.08.2013, p. 3918-3925.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A maximum-margin genetic algorithm for misclassification cost minimizing feature selection problem

AU - Pendharkar, Parag C.

PY - 2013/8/1

Y1 - 2013/8/1

N2 - We consider a feature selection problem where the decision-making objective is to minimize overall misclassification cost by selecting relevant features from a training dataset. We propose a two-stage solution approach for solving misclassification cost minimizing feature selection (MCMFS) problem. Additionally, we propose a maximum-margin genetic algorithm (MMGA) that maximizes margin of separation between classes by taking into account all examples as opposed to maximizing margin of separation using a few support vectors. Feature selection is carried out by either an exhaustive or a heuristic simulated annealing approach in the first stage and a cost sensitive classification using either MMGA or cost sensitive support vector machines (SVM) in the second stage. Using simulated and real-world data sets and different misclassification cost matrices, we test our two-stage approach for solving the MCMFS problem. Our results indicate that feature selection plays an important role when misclassification cost asymmetries increase and the MMGA shows equal or better performance than the SVM.

AB - We consider a feature selection problem where the decision-making objective is to minimize overall misclassification cost by selecting relevant features from a training dataset. We propose a two-stage solution approach for solving misclassification cost minimizing feature selection (MCMFS) problem. Additionally, we propose a maximum-margin genetic algorithm (MMGA) that maximizes margin of separation between classes by taking into account all examples as opposed to maximizing margin of separation using a few support vectors. Feature selection is carried out by either an exhaustive or a heuristic simulated annealing approach in the first stage and a cost sensitive classification using either MMGA or cost sensitive support vector machines (SVM) in the second stage. Using simulated and real-world data sets and different misclassification cost matrices, we test our two-stage approach for solving the MCMFS problem. Our results indicate that feature selection plays an important role when misclassification cost asymmetries increase and the MMGA shows equal or better performance than the SVM.

UR - http://www.scopus.com/inward/record.url?scp=84875363688&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84875363688&partnerID=8YFLogxK

U2 - 10.1016/j.eswa.2012.12.086

DO - 10.1016/j.eswa.2012.12.086

M3 - Article

AN - SCOPUS:84875363688

VL - 40

SP - 3918

EP - 3925

JO - Expert Systems with Applications

JF - Expert Systems with Applications

SN - 0957-4174

IS - 10

ER -