Towards crafting an improved functional link artificial neural network based on differential evolution and feature selection

Sanjeev Kumar Dash, Ajit Kumar Behera, Satchidananda Dehuri, Sung-Bae Cho, Gi Nam Wang

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

The proposed work describes an improved functional link artificial neural network (FLANN) for classification. The improvement in terms of classification accuracy of the network is realized through differential evolution (DE) and filter based feature selection approach. Information gain theory is used to filter out irrelevant features and provide relevant features to the functional expansion unit of FLANN as an input, which in turn maps low to high dimensional feature space for constructing an improved classifier. To fine tune the weight vector of the given network, differential evolution is used. The work is validated using skewed and balanced dataset retrieved from the University of California Irvine (UCI) repository. Our systematic experimental study divulges that the performance of the differential-evolution trained FLANN is promising than genetic algorithm trained FLANN, ISO-FLANN, and PSO-BP.

Original languageEnglish
Pages (from-to)195-208
Number of pages14
JournalInformatica (Slovenia)
Volume39
Issue number2
Publication statusPublished - 2015 Jan 1

Fingerprint

Differential Evolution
Feature Selection
Artificial Neural Network
Feature extraction
Neural networks
Filter
Information Gain
Feature Space
Repository
Particle swarm optimization (PSO)
Experimental Study
High-dimensional
Classifiers
Genetic algorithms
Classifier
Genetic Algorithm
Unit

All Science Journal Classification (ASJC) codes

  • Software
  • Theoretical Computer Science
  • Computer Science Applications
  • Artificial Intelligence

Cite this

Dash, Sanjeev Kumar ; Behera, Ajit Kumar ; Dehuri, Satchidananda ; Cho, Sung-Bae ; Wang, Gi Nam. / Towards crafting an improved functional link artificial neural network based on differential evolution and feature selection. In: Informatica (Slovenia). 2015 ; Vol. 39, No. 2. pp. 195-208.
@article{abbf5723504b42c4af2c54260ca74a7c,
title = "Towards crafting an improved functional link artificial neural network based on differential evolution and feature selection",
abstract = "The proposed work describes an improved functional link artificial neural network (FLANN) for classification. The improvement in terms of classification accuracy of the network is realized through differential evolution (DE) and filter based feature selection approach. Information gain theory is used to filter out irrelevant features and provide relevant features to the functional expansion unit of FLANN as an input, which in turn maps low to high dimensional feature space for constructing an improved classifier. To fine tune the weight vector of the given network, differential evolution is used. The work is validated using skewed and balanced dataset retrieved from the University of California Irvine (UCI) repository. Our systematic experimental study divulges that the performance of the differential-evolution trained FLANN is promising than genetic algorithm trained FLANN, ISO-FLANN, and PSO-BP.",
author = "Dash, {Sanjeev Kumar} and Behera, {Ajit Kumar} and Satchidananda Dehuri and Sung-Bae Cho and Wang, {Gi Nam}",
year = "2015",
month = "1",
day = "1",
language = "English",
volume = "39",
pages = "195--208",
journal = "Informatica",
issn = "0350-5596",
publisher = "Slovene Society Informatika",
number = "2",

}

Towards crafting an improved functional link artificial neural network based on differential evolution and feature selection. / Dash, Sanjeev Kumar; Behera, Ajit Kumar; Dehuri, Satchidananda; Cho, Sung-Bae; Wang, Gi Nam.

In: Informatica (Slovenia), Vol. 39, No. 2, 01.01.2015, p. 195-208.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Towards crafting an improved functional link artificial neural network based on differential evolution and feature selection

AU - Dash, Sanjeev Kumar

AU - Behera, Ajit Kumar

AU - Dehuri, Satchidananda

AU - Cho, Sung-Bae

AU - Wang, Gi Nam

PY - 2015/1/1

Y1 - 2015/1/1

N2 - The proposed work describes an improved functional link artificial neural network (FLANN) for classification. The improvement in terms of classification accuracy of the network is realized through differential evolution (DE) and filter based feature selection approach. Information gain theory is used to filter out irrelevant features and provide relevant features to the functional expansion unit of FLANN as an input, which in turn maps low to high dimensional feature space for constructing an improved classifier. To fine tune the weight vector of the given network, differential evolution is used. The work is validated using skewed and balanced dataset retrieved from the University of California Irvine (UCI) repository. Our systematic experimental study divulges that the performance of the differential-evolution trained FLANN is promising than genetic algorithm trained FLANN, ISO-FLANN, and PSO-BP.

AB - The proposed work describes an improved functional link artificial neural network (FLANN) for classification. The improvement in terms of classification accuracy of the network is realized through differential evolution (DE) and filter based feature selection approach. Information gain theory is used to filter out irrelevant features and provide relevant features to the functional expansion unit of FLANN as an input, which in turn maps low to high dimensional feature space for constructing an improved classifier. To fine tune the weight vector of the given network, differential evolution is used. The work is validated using skewed and balanced dataset retrieved from the University of California Irvine (UCI) repository. Our systematic experimental study divulges that the performance of the differential-evolution trained FLANN is promising than genetic algorithm trained FLANN, ISO-FLANN, and PSO-BP.

UR - http://www.scopus.com/inward/record.url?scp=84936992624&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84936992624&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:84936992624

VL - 39

SP - 195

EP - 208

JO - Informatica

JF - Informatica

SN - 0350-5596

IS - 2

ER -