Learning Functional Dependency Networks Based on
Genetic Programming
W. Shum, K. Leung, and M. Wong. Proceedings of the 5th IEEE International Conference
on Data Mining (ICDM 2005), page 394--401. Houston, Texas, USA, IEEE Computer Society, (27-30 November 2005)
Abstract
Bayesian Network (BN) is a powerful network model,
which represents a set of variables in the domain and
provides the probabilistic relationships among them.
But BN can handle discrete values only; it cannot
handle continuous, interval and ordinal ones, which
must be converted to discrete values and the order
information is lost. Thus, BN tends to have higher
network complexity and lower understandability. In this
paper, we present a novel dependency network which can
handle discrete, continuous, interval and ordinal
values through functions; it has lower network
complexity and stronger expressive power; it can
represent any kind of relationships; and it can
incorporate a-priori knowledge though user-defined
functions. We also propose a novel Genetic Programming
(GP) to learn dependency networks. The novel GP does
not use any knowledge-guided nor application-oriented
operator, thus it is robust and easy to replicate. The
experimental results demonstrate that the novel GP can
successfully discover the target novel dependency
networks, which have the highest accuracy and the
lowest network complexity.
%0 Conference Paper
%1 conf/icdm/ShumLW05
%A Shum, Wing-Ho
%A Leung, Kwong-Sak
%A Wong, Man Leung
%B Proceedings of the 5th IEEE International Conference
on Data Mining (ICDM 2005)
%C Houston, Texas, USA
%D 2005
%I IEEE Computer Society
%K algorithms, genetic programming
%P 394--401
%T Learning Functional Dependency Networks Based on
Genetic Programming
%U http://doi.ieeecomputersociety.org/10.1109/ICDM.2005.86
%X Bayesian Network (BN) is a powerful network model,
which represents a set of variables in the domain and
provides the probabilistic relationships among them.
But BN can handle discrete values only; it cannot
handle continuous, interval and ordinal ones, which
must be converted to discrete values and the order
information is lost. Thus, BN tends to have higher
network complexity and lower understandability. In this
paper, we present a novel dependency network which can
handle discrete, continuous, interval and ordinal
values through functions; it has lower network
complexity and stronger expressive power; it can
represent any kind of relationships; and it can
incorporate a-priori knowledge though user-defined
functions. We also propose a novel Genetic Programming
(GP) to learn dependency networks. The novel GP does
not use any knowledge-guided nor application-oriented
operator, thus it is robust and easy to replicate. The
experimental results demonstrate that the novel GP can
successfully discover the target novel dependency
networks, which have the highest accuracy and the
lowest network complexity.
%@ 0-7695-2278-5
@inproceedings{conf/icdm/ShumLW05,
abstract = {Bayesian Network (BN) is a powerful network model,
which represents a set of variables in the domain and
provides the probabilistic relationships among them.
But BN can handle discrete values only; it cannot
handle continuous, interval and ordinal ones, which
must be converted to discrete values and the order
information is lost. Thus, BN tends to have higher
network complexity and lower understandability. In this
paper, we present a novel dependency network which can
handle discrete, continuous, interval and ordinal
values through functions; it has lower network
complexity and stronger expressive power; it can
represent any kind of relationships; and it can
incorporate a-priori knowledge though user-defined
functions. We also propose a novel Genetic Programming
(GP) to learn dependency networks. The novel GP does
not use any knowledge-guided nor application-oriented
operator, thus it is robust and easy to replicate. The
experimental results demonstrate that the novel GP can
successfully discover the target novel dependency
networks, which have the highest accuracy and the
lowest network complexity.},
added-at = {2008-06-19T17:46:40.000+0200},
address = {Houston, Texas, USA},
author = {Shum, Wing-Ho and Leung, Kwong-Sak and Wong, Man Leung},
bibdate = {2005-12-21},
bibsource = {DBLP,
http://dblp.uni-trier.de/db/conf/icdm/icdm2005.html#ShumLW05},
biburl = {https://www.bibsonomy.org/bibtex/22461bee242feaaf143fb7b19fdff10ad/brazovayeye},
booktitle = {Proceedings of the 5th IEEE International Conference
on Data Mining (ICDM 2005)},
interhash = {8ce57ff29f33962c6d3da7df6fa9e035},
intrahash = {2461bee242feaaf143fb7b19fdff10ad},
isbn = {0-7695-2278-5},
keywords = {algorithms, genetic programming},
month = {27-30 November},
pages = {394--401},
publisher = {IEEE Computer Society},
timestamp = {2008-06-19T17:51:38.000+0200},
title = {Learning Functional Dependency Networks Based on
Genetic Programming},
url = {http://doi.ieeecomputersociety.org/10.1109/ICDM.2005.86},
year = 2005
}