Selecting features in neurofuzzy modelling by multi-objective genetic algorithms

Emmanouilidis, Christos and Hunter, Andrew and MacIntyre, John and Cox, Chris (1999) Selecting features in neurofuzzy modelling by multi-objective genetic algorithms. In: ICANN 1999, 7-10 September 1999, Edinburgh.

Documents
Selecting features in neurofuzzy modelling by multi-objective genetic algorithms
Selecting features in neurofuzzy modelling by multi-objective genetic algorithms
[img]
[Download]
Request a copy
[img] PDF
CE_ICANN.pdf
Restricted to Repository staff only

42kB

Full text URL: http://anc.ed.ac.uk/ICANN99/

Abstract

ABSTRACT
Empirical modelling in high dimensional spaces is usually preceded by a feature selection stage. Irrelevant or noisy features unnecessarily increase the complexity of the
problem and can degrade modelling performance. Here, multiobjective genetic algorithms, are proposed, as effective means of evolving a diverse population of alternative feature sets with various accuracy/complexity
trade-offs. They are shown to be particularly successful in neurofuzzy modelling, in conjunction with a method for performing fast fitness evaluation. The major contributions of this paper are in the use of a specific type of
multiobjective genetic algorithm, based on the concept of dominance, for feature selection; and the combination of fast fitness evaluation of neurofuzzy models with a genetic algorithm. The effectiveness of the proposed approach is
demonstrated on two high dimensional regression problems.

Item Type:Conference or Workshop Item (Paper)
Additional Information:ABSTRACT Empirical modelling in high dimensional spaces is usually preceded by a feature selection stage. Irrelevant or noisy features unnecessarily increase the complexity of the problem and can degrade modelling performance. Here, multiobjective genetic algorithms, are proposed, as effective means of evolving a diverse population of alternative feature sets with various accuracy/complexity trade-offs. They are shown to be particularly successful in neurofuzzy modelling, in conjunction with a method for performing fast fitness evaluation. The major contributions of this paper are in the use of a specific type of multiobjective genetic algorithm, based on the concept of dominance, for feature selection; and the combination of fast fitness evaluation of neurofuzzy models with a genetic algorithm. The effectiveness of the proposed approach is demonstrated on two high dimensional regression problems.
Keywords:Genetic algorithms, Algorithms, Neurofuzzy
Subjects:G Mathematical and Computer Sciences > G760 Machine Learning
Divisions:College of Science > School of Computer Science
ID Code:1898
Deposited By: Tammie Farley
Deposited On:25 Jun 2009 13:31
Last Modified:13 Mar 2013 08:32

Repository Staff Only: item control page