Learning functional dependency networks based on genetic programming

Wing Ho SHUM, Kwong Sak LEUNG, Man Leung WONG

Research output: Book Chapters | Papers in Conference ProceedingsConference paper (refereed)Researchpeer-review

1 Scopus Citations

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.
Original languageEnglish
Title of host publicationProceedings of the 5th IEEE International Conference on Data Mining, ICDM 2005
PublisherIEEE Computer Society
Pages394-401
Number of pages8
DOIs
Publication statusPublished - 1 Jan 2005

Fingerprint

Genetic programming
Bayesian networks

Bibliographical note

Paper presented at the 5th IEEE International Conference on Data Mining (ICDM 2005), 27-30 November 2005, Houston, Texas.
ISBN of the source publication: 9780769522784

Cite this

SHUM, W. H., LEUNG, K. S., & WONG, M. L. (2005). Learning functional dependency networks based on genetic programming. In Proceedings of the 5th IEEE International Conference on Data Mining, ICDM 2005 (pp. 394-401). IEEE Computer Society. https://doi.org/10.1109/ICDM.2005.86
SHUM, Wing Ho ; LEUNG, Kwong Sak ; WONG, Man Leung. / Learning functional dependency networks based on genetic programming. Proceedings of the 5th IEEE International Conference on Data Mining, ICDM 2005. IEEE Computer Society, 2005. pp. 394-401
@inproceedings{408e1387fbfe479d903b923eaa57b818,
title = "Learning functional dependency networks based on genetic programming",
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.",
author = "SHUM, {Wing Ho} and LEUNG, {Kwong Sak} and WONG, {Man Leung}",
note = "Paper presented at the 5th IEEE International Conference on Data Mining (ICDM 2005), 27-30 November 2005, Houston, Texas. ISBN of the source publication: 9780769522784",
year = "2005",
month = "1",
day = "1",
doi = "10.1109/ICDM.2005.86",
language = "English",
pages = "394--401",
booktitle = "Proceedings of the 5th IEEE International Conference on Data Mining, ICDM 2005",
publisher = "IEEE Computer Society",
address = "United States",

}

SHUM, WH, LEUNG, KS & WONG, ML 2005, Learning functional dependency networks based on genetic programming. in Proceedings of the 5th IEEE International Conference on Data Mining, ICDM 2005. IEEE Computer Society, pp. 394-401. https://doi.org/10.1109/ICDM.2005.86

Learning functional dependency networks based on genetic programming. / SHUM, Wing Ho; LEUNG, Kwong Sak; WONG, Man Leung.

Proceedings of the 5th IEEE International Conference on Data Mining, ICDM 2005. IEEE Computer Society, 2005. p. 394-401.

Research output: Book Chapters | Papers in Conference ProceedingsConference paper (refereed)Researchpeer-review

TY - GEN

T1 - Learning functional dependency networks based on genetic programming

AU - SHUM, Wing Ho

AU - LEUNG, Kwong Sak

AU - WONG, Man Leung

N1 - Paper presented at the 5th IEEE International Conference on Data Mining (ICDM 2005), 27-30 November 2005, Houston, Texas. ISBN of the source publication: 9780769522784

PY - 2005/1/1

Y1 - 2005/1/1

N2 - 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.

AB - 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.

UR - http://commons.ln.edu.hk/sw_master/7102

U2 - 10.1109/ICDM.2005.86

DO - 10.1109/ICDM.2005.86

M3 - Conference paper (refereed)

SP - 394

EP - 401

BT - Proceedings of the 5th IEEE International Conference on Data Mining, ICDM 2005

PB - IEEE Computer Society

ER -

SHUM WH, LEUNG KS, WONG ML. Learning functional dependency networks based on genetic programming. In Proceedings of the 5th IEEE International Conference on Data Mining, ICDM 2005. IEEE Computer Society. 2005. p. 394-401 https://doi.org/10.1109/ICDM.2005.86