One-Class Learning with Multi-Objective Genetic
Programming
R. Curry, and M. Heywood. Proceedings of the IEEE International Conference on
Systems, Man, and Cybernetics, page 1938--1945. Montreal, IEEE Press, (7-10 October 2007)
Abstract
One-class classification naturally only provides one
class of exemplars on which to construct the
classification model. In this work, multi-objective
genetic programming (GP) allows the one-class learning
problem to be decomposed by multiple GP classifiers,
each attempting to identify only a subset of the target
data to classify. In order for GP to identify
appropriate subsets of the one-class data, artificial
outclass data is generated in and around the provided
inclass data. A local Gaussian wrapper is employed
where this reinforces a novelty detection as opposed to
a discrimination approach to classification.
Furthermore, a hierarchical subset selection strategy
is used to deal with the necessarily large number of
generated outclass exemplars. The proposed approach is
demonstrated on three one-class classification datasets
and was found to be competitive with a one-class SVM
classifier and a binary SVM classifier.
%0 Conference Paper
%1 Curry:2007:SMCb
%A Curry, R.
%A Heywood, M. I.
%B Proceedings of the IEEE International Conference on
Systems, Man, and Cybernetics
%C Montreal
%D 2007
%I IEEE Press
%K algorithms, evolutionary genetic learning multi-criteria one-class optimisation, programming,
%P 1938--1945
%T One-Class Learning with Multi-Objective Genetic
Programming
%U http://users.cs.dal.ca/~mheywood/X-files/Publications/rcurry_SMC07.pdf
%X One-class classification naturally only provides one
class of exemplars on which to construct the
classification model. In this work, multi-objective
genetic programming (GP) allows the one-class learning
problem to be decomposed by multiple GP classifiers,
each attempting to identify only a subset of the target
data to classify. In order for GP to identify
appropriate subsets of the one-class data, artificial
outclass data is generated in and around the provided
inclass data. A local Gaussian wrapper is employed
where this reinforces a novelty detection as opposed to
a discrimination approach to classification.
Furthermore, a hierarchical subset selection strategy
is used to deal with the necessarily large number of
generated outclass exemplars. The proposed approach is
demonstrated on three one-class classification datasets
and was found to be competitive with a one-class SVM
classifier and a binary SVM classifier.
%@ 1-4244-0991-8
@inproceedings{Curry:2007:SMCb,
abstract = {One-class classification naturally only provides one
class of exemplars on which to construct the
classification model. In this work, multi-objective
genetic programming (GP) allows the one-class learning
problem to be decomposed by multiple GP classifiers,
each attempting to identify only a subset of the target
data to classify. In order for GP to identify
appropriate subsets of the one-class data, artificial
outclass data is generated in and around the provided
inclass data. A local Gaussian wrapper is employed
where this reinforces a novelty detection as opposed to
a discrimination approach to classification.
Furthermore, a hierarchical subset selection strategy
is used to deal with the necessarily large number of
generated outclass exemplars. The proposed approach is
demonstrated on three one-class classification datasets
and was found to be competitive with a one-class SVM
classifier and a binary SVM classifier.},
added-at = {2008-06-19T17:35:00.000+0200},
address = {Montreal},
author = {Curry, R. and Heywood, M. I.},
biburl = {https://www.bibsonomy.org/bibtex/2e56e393f984f67efbef363217988603c/brazovayeye},
booktitle = {Proceedings of the IEEE International Conference on
Systems, Man, and Cybernetics},
interhash = {63de2c76c03312b01d79c7538e283697},
intrahash = {e56e393f984f67efbef363217988603c},
isbn = {1-4244-0991-8},
keywords = {algorithms, evolutionary genetic learning multi-criteria one-class optimisation, programming,},
month = {7-10 October},
notes = {http://www.smc2007.org/program.html
rcurry_SMC07.pdf is twenty pages},
pages = {1938--1945},
publisher = {IEEE Press},
timestamp = {2008-06-19T17:38:18.000+0200},
title = {One-Class Learning with Multi-Objective Genetic
Programming},
url = {http://users.cs.dal.ca/~mheywood/X-files/Publications/rcurry_SMC07.pdf},
year = 2007
}