Feature selection using genetic algorithms and probabilistic neural networks

Hunter, Andrew (2000) Feature selection using genetic algorithms and probabilistic neural networks. Neural Computing & Applications, 9 (2). pp. 124-132. ISSN 0941-0643

Documents
Feature selection using genetic algorithms and probabilistic neural networks
Feature Selection using Genetic Algorithms and Probabilistic Neural Networks
[img]
[Download]
[img]
Preview
PDF
FeatureSelectionGAPNN.pdf - Whole Document

86kB

Full text URL: http://dx.doi.org/10.1007/s005210070023

Abstract

Selection of input variables is a key stage in building
predictive models, and an important form of data mining. As exhaustive evaluation of potential input sets using full non-linear models is impractical, it is necessary to use simple fast-evaluating models and heuristic selection strategies. This paper discusses a fast, efficient, and powerful nonlinear input selection procedure using a combination of Probabilistic Neural Networks and repeated
bitwise gradient descent. The algorithm is compared
with forward elimination, backward elimination and genetic algorithms using a selection of real-world data sets. The algorithm has comparative performance and greatly reduced execution time with respect to these alternative approaches. It is demonstrated empirically that reliable results cannot be gained using any of these approaches without the use of resampling.

Item Type:Article
Additional Information:Selection of input variables is a key stage in building predictive models, and an important form of data mining. As exhaustive evaluation of potential input sets using full non-linear models is impractical, it is necessary to use simple fast-evaluating models and heuristic selection strategies. This paper discusses a fast, efficient, and powerful nonlinear input selection procedure using a combination of Probabilistic Neural Networks and repeated bitwise gradient descent. The algorithm is compared with forward elimination, backward elimination and genetic algorithms using a selection of real-world data sets. The algorithm has comparative performance and greatly reduced execution time with respect to these alternative approaches. It is demonstrated empirically that reliable results cannot be gained using any of these approaches without the use of resampling.
Keywords:Probabalistic neural networks, Feature selection, Genetic algorithms
Subjects:G Mathematical and Computer Sciences > G700 Artificial Intelligence
G Mathematical and Computer Sciences > G760 Machine Learning
G Mathematical and Computer Sciences > G730 Neural Computing
Divisions:College of Science > School of Computer Science
ID Code:1913
Deposited By: Tammie Farley
Deposited On:29 Jan 2010 12:00
Last Modified:13 Mar 2013 08:33

Repository Staff Only: item control page