A comparison of genetic programming variants for data
classification
J. Eggermont, A. Eiben, and J. van
Hemert. Advances in Intelligent Data Analysis, Third
International Symposium, IDA-99, volume 1642 of LNCS, page 281--290. Amsterdam, The Netherlands, Springer-Verlag, (9--11 August 1999)
Abstract
We report a comparative study on different variations
of genetic programming applied on binary data
classification problems. The first genetic programming
variant is weighting data records for calculating the
classification error and modifying the weights during
the run. Hereby the algorithm is defining its own
fitness function in an on-line fashion giving higher
weights to `hard' records. Another novel feature we
study is the atomic representation, where
`Booleanization' of data is not performed at the root,
but at the leafs of the trees and only Boolean
functions are used in the trees' body. As a third
aspect we look at generational and steady-state models
in combination of both features.
%0 Conference Paper
%1 EEH99b
%A Eggermont, Jeroen
%A Eiben, Agoston E.
%A van
Hemert, Jano I.
%B Advances in Intelligent Data Analysis, Third
International Symposium, IDA-99
%C Amsterdam, The Netherlands
%D 1999
%E Hand, David J.
%E Kok, Joost N.
%E Berthold, Michael R.
%I Springer-Verlag
%K algorithms, classification, data genetic mining programming,
%P 281--290
%T A comparison of genetic programming variants for data
classification
%U http://www.vanhemert.co.uk/publications/ida99.A_comparison_of_genetic_programming_variants_for_data_classification.ps.gz
%V 1642
%X We report a comparative study on different variations
of genetic programming applied on binary data
classification problems. The first genetic programming
variant is weighting data records for calculating the
classification error and modifying the weights during
the run. Hereby the algorithm is defining its own
fitness function in an on-line fashion giving higher
weights to `hard' records. Another novel feature we
study is the atomic representation, where
`Booleanization' of data is not performed at the root,
but at the leafs of the trees and only Boolean
functions are used in the trees' body. As a third
aspect we look at generational and steady-state models
in combination of both features.
%@ 3-540-66332-0
@inproceedings{EEH99b,
abstract = {We report a comparative study on different variations
of genetic programming applied on binary data
classification problems. The first genetic programming
variant is weighting data records for calculating the
classification error and modifying the weights during
the run. Hereby the algorithm is defining its own
fitness function in an on-line fashion giving higher
weights to `hard' records. Another novel feature we
study is the atomic representation, where
`Booleanization' of data is not performed at the root,
but at the leafs of the trees and only Boolean
functions are used in the trees' body. As a third
aspect we look at generational and steady-state models
in combination of both features.},
added-at = {2008-06-19T17:35:00.000+0200},
address = {Amsterdam, The Netherlands},
author = {Eggermont, Jeroen and Eiben, Agoston E. and {van
Hemert}, Jano I.},
biburl = {https://www.bibsonomy.org/bibtex/22a6ff5aa4a14b60668add91cc7eee70e/brazovayeye},
booktitle = {Advances in Intelligent Data Analysis, Third
International Symposium, IDA-99},
editor = {Hand, David J. and Kok, Joost N. and Berthold, Michael R.},
email = {jvhemert@cs.leidenuniv.nl},
interhash = {8a55488ca35a07a806c5813c25e29abc},
intrahash = {2a6ff5aa4a14b60668add91cc7eee70e},
isbn = {3-540-66332-0},
keywords = {algorithms, classification, data genetic mining programming,},
month = {9--11 August},
notes = {IDA-99, Booleanization of inputs, ML: Australian
credit, German Credit, Heart Disease, Pima. steady
state. SAW-ing},
pages = {281--290},
publisher = {Springer-Verlag},
publisher_address = {Berlin},
series = {LNCS},
timestamp = {2008-06-19T17:39:07.000+0200},
title = {A comparison of genetic programming variants for data
classification},
url = {http://www.vanhemert.co.uk/publications/ida99.A_comparison_of_genetic_programming_variants_for_data_classification.ps.gz},
volume = 1642,
year = 1999
}