Co-operative evolution of a neural classifier and feature subset

Jennifer Hallinan, Paul Jackway

Research output: Chapter in Book/Report/Conference proceedingChapterpeer-review

5 Citations (Scopus)

Abstract

This paper describes a novel feature selection algorithm which utilizes a genetic algorithm to select a feature subset in conjunction with the weights for a three-layer feedforward network classifier. The algorithm was tested on the \ionosphere" data set from UC Irvine, and on an artifically generated data set. This approach produces results comparable to those reported for other algorithms on the ionosphere data, but using fewer input features and a simpler neural network architecture. These results indicate that tailoring a neural network classifier to a specific subset of features has the potential to produce a classifier with low classification error and good generalizability.

Original languageEnglish
Title of host publicationSimulated Evolution and Learning
Subtitle of host publicationSecond Asia-Pacific Conference on Simulated Evolution and Learning, SEAL’98 Canberra, Australia, November 24–27, 1998 Selected Papers
EditorsBob McKay, Xin Yao, Charles S. Newton, Jong-Hwan Kim, Takeshi Furuhashi
Place of PublicationBerlin
PublisherSpringer, Springer Nature
Pages397-404
Number of pages8
Volume1585
ISBN (Electronic)9783540488736
ISBN (Print)9783540659075
DOIs
Publication statusPublished - 1999
Externally publishedYes
Event2nd Asia-Pacific Conference on Simulated Evolution and Learning, SEAL 1998 - Canberra, Australia
Duration: 24 Nov 199827 Nov 1998

Publication series

NameLecture Notes in Computer Science
PublisherSpringer Berlin Heidelberg
Volume1585
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other2nd Asia-Pacific Conference on Simulated Evolution and Learning, SEAL 1998
CountryAustralia
CityCanberra
Period24/11/9827/11/98

Keywords

  • Classification
  • Genetic algorithm
  • Ionosphere
  • Neural network

Fingerprint

Dive into the research topics of 'Co-operative evolution of a neural classifier and feature subset'. Together they form a unique fingerprint.

Cite this