Emmanouilidis, Christos and Hunter, Andrew and MacIntyre, John (2000) A multiobjective evolutionary setting for feature selection and a commonality-based crossover operator. In: 2000 Congress on Evolutionary Computation, 16th-19th July 2000, La Jolla, CA, USA.

PDF CEC.pdf Restricted to Repository staff only Download (90Kb) | Request a copy

Abstract

Abstract- Feature subset selection is a common and key problem in many classification and regression tasks. It can be viewed as a multi-objective optimisation problem, since, in the simplest case, it involves feature subset size minimisation and performance maximisation. Here, a multiobjective evolutionary approach is proposed for feature selection. A novel commonality-based crossover operator is introduced and placed in the multiobjective evolutionary setting. This specialised operator helps to preserve building blocks with promising performance. The multiobjective evolutionary algothim employs the novel crossover operator in order to evolve a diverse population of feature subsets with different subset size/performance trade-offs. Selection bias reduction is achieved by means of resampling. We argue that this is a generic approach, which can be used in many modelling problems. It is applied to feature selection on different neural network architectures. Results from experiments with high dimensional benchmarking data sets are given.

Item Type:	Conference or Workshop Item (Paper)
Additional Information:	Abstract- Feature subset selection is a common and key problem in many classification and regression tasks. It can be viewed as a multi-objective optimisation problem, since, in the simplest case, it involves feature subset size minimisation and performance maximisation. Here, a multiobjective evolutionary approach is proposed for feature selection. A novel commonality-based crossover operator is introduced and placed in the multiobjective evolutionary setting. This specialised operator helps to preserve building blocks with promising performance. The multiobjective evolutionary algothim employs the novel crossover operator in order to evolve a diverse population of feature subsets with different subset size/performance trade-offs. Selection bias reduction is achieved by means of resampling. We argue that this is a generic approach, which can be used in many modelling problems. It is applied to feature selection on different neural network architectures. Results from experiments with high dimensional benchmarking data sets are given.
Keywords:	Algorithms, Feature selection, Multiobjective evolutionary algorithm, Neural networks
Subjects:	G Mathematical and Computer Sciences > G730 Neural Computing
Divisions:	College of Sciences > Faculty of Science > Lincoln School of Computer Science
Depositing User:	Tammie Farley
Date Deposited:	23 Jun 2009 14:55
Last Modified:	13 Mar 2013 08:32
URI:	http://eprints.lincoln.ac.uk/id/eprint/1907

Actions (login required)

View Item