Masters Theses

A hybrid approach for feature subset selection using ant colony optimization and artificial neural networks

Abstract

"Feature selection deals with selecting a subset of feature from a data set to predict the output with an acceptable level of accuracy. Feature selection problems have been solved previously by researchers using various meta-heuristic algorithms like branch and bound method, genetic algorithm, simulated annealing etc. This thesis presents a hybrid approach using artificial neural network and ant colony optimization, which would find out the inter-variable relationship amongst a subset of feature, if any, to predict the output accurately"--Abstract, page iv.

Advisor(s)

Ramakrishnan, Sreeram

Committee Member(s)

Dagli, Cihan H., 1949-
Enke, David Lee, 1965-

Department(s)

Engineering Management and Systems Engineering

Degree Name

M.S. in Systems Engineering

Publisher

University of Missouri--Rolla

Publication Date

Spring 2007

Journal article titles appearing in thesis/dissertation

  • Modified hybrid approach for feature selection using ant colony optimization and neural networks

Pagination

xii, 96 pages

Rights

© 2007 Rahul Karthik Sivagaminathan, All rights reserved.

Document Type

Thesis - Citation

File Type

text

Language

English

Subject Headings

Ants -- Behavior -- Mathematical modelsMathematical optimizationNeural networks (Computer science)

Thesis Number

T 9145

Print OCLC #

173405216

This document is currently not available here.

Share My Thesis If you are the author of this work and would like to grant permission to make it openly accessible to all, please click the button above.

Share

 
COinS