A novel hybrid evolutionary algorithm for learning Bayesian networks from incomplete data

Yuan Yuan GUO, Man Leung WONG, Zhi Hua CAI

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

3 Citations (Scopus)

Abstract

Existing Structural Expectation-Maximization (EM) algorithms for learning Bayesian networks from incomplete data usually adopt the greedy hill climbing search method, which may make the algorithms find sub-optimal solutions. In this paper, we present a new Structural EM algorithm which employs a hybrid evolutionary algorithm as the search method. The experimental results on the data sets generated from several benchmark networks illustrate that our algorithm outperforms some state-of-the-art learning algorithms.
Original languageEnglish
Title of host publicationProceedings of the 2006 IEEE Congress on Evolutionary Computation, CEC 2006
PublisherInstitute of Electrical and Electronics Engineers
Pages916-923
Number of pages8
DOIs
Publication statusPublished - 1 Jan 2006

Fingerprint

Bayesian networks
Evolutionary algorithms
Learning algorithms

Bibliographical note

Paper presented at the 2006 IEEE Congress on Evolutionary Computation (CEC 2006), 16-21 July 2006, Vancouver, Canada.
ISBN of the source publication: 9780780394872

Cite this

GUO, Y. Y., WONG, M. L., & CAI, Z. H. (2006). A novel hybrid evolutionary algorithm for learning Bayesian networks from incomplete data. In Proceedings of the 2006 IEEE Congress on Evolutionary Computation, CEC 2006 (pp. 916-923). Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/CEC.2006.1688409
GUO, Yuan Yuan ; WONG, Man Leung ; CAI, Zhi Hua. / A novel hybrid evolutionary algorithm for learning Bayesian networks from incomplete data. Proceedings of the 2006 IEEE Congress on Evolutionary Computation, CEC 2006. Institute of Electrical and Electronics Engineers, 2006. pp. 916-923
@inproceedings{dff5ffd89a824c92b79741aa9a6bc48b,
title = "A novel hybrid evolutionary algorithm for learning Bayesian networks from incomplete data",
abstract = "Existing Structural Expectation-Maximization (EM) algorithms for learning Bayesian networks from incomplete data usually adopt the greedy hill climbing search method, which may make the algorithms find sub-optimal solutions. In this paper, we present a new Structural EM algorithm which employs a hybrid evolutionary algorithm as the search method. The experimental results on the data sets generated from several benchmark networks illustrate that our algorithm outperforms some state-of-the-art learning algorithms.",
author = "GUO, {Yuan Yuan} and WONG, {Man Leung} and CAI, {Zhi Hua}",
note = "Paper presented at the 2006 IEEE Congress on Evolutionary Computation (CEC 2006), 16-21 July 2006, Vancouver, Canada. ISBN of the source publication: 9780780394872",
year = "2006",
month = "1",
day = "1",
doi = "10.1109/CEC.2006.1688409",
language = "English",
pages = "916--923",
booktitle = "Proceedings of the 2006 IEEE Congress on Evolutionary Computation, CEC 2006",
publisher = "Institute of Electrical and Electronics Engineers",

}

GUO, YY, WONG, ML & CAI, ZH 2006, A novel hybrid evolutionary algorithm for learning Bayesian networks from incomplete data. in Proceedings of the 2006 IEEE Congress on Evolutionary Computation, CEC 2006. Institute of Electrical and Electronics Engineers, pp. 916-923. https://doi.org/10.1109/CEC.2006.1688409

A novel hybrid evolutionary algorithm for learning Bayesian networks from incomplete data. / GUO, Yuan Yuan; WONG, Man Leung; CAI, Zhi Hua.

Proceedings of the 2006 IEEE Congress on Evolutionary Computation, CEC 2006. Institute of Electrical and Electronics Engineers, 2006. p. 916-923.

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

TY - GEN

T1 - A novel hybrid evolutionary algorithm for learning Bayesian networks from incomplete data

AU - GUO, Yuan Yuan

AU - WONG, Man Leung

AU - CAI, Zhi Hua

N1 - Paper presented at the 2006 IEEE Congress on Evolutionary Computation (CEC 2006), 16-21 July 2006, Vancouver, Canada. ISBN of the source publication: 9780780394872

PY - 2006/1/1

Y1 - 2006/1/1

N2 - Existing Structural Expectation-Maximization (EM) algorithms for learning Bayesian networks from incomplete data usually adopt the greedy hill climbing search method, which may make the algorithms find sub-optimal solutions. In this paper, we present a new Structural EM algorithm which employs a hybrid evolutionary algorithm as the search method. The experimental results on the data sets generated from several benchmark networks illustrate that our algorithm outperforms some state-of-the-art learning algorithms.

AB - Existing Structural Expectation-Maximization (EM) algorithms for learning Bayesian networks from incomplete data usually adopt the greedy hill climbing search method, which may make the algorithms find sub-optimal solutions. In this paper, we present a new Structural EM algorithm which employs a hybrid evolutionary algorithm as the search method. The experimental results on the data sets generated from several benchmark networks illustrate that our algorithm outperforms some state-of-the-art learning algorithms.

UR - http://commons.ln.edu.hk/sw_master/7077

U2 - 10.1109/CEC.2006.1688409

DO - 10.1109/CEC.2006.1688409

M3 - Conference paper (refereed)

SP - 916

EP - 923

BT - Proceedings of the 2006 IEEE Congress on Evolutionary Computation, CEC 2006

PB - Institute of Electrical and Electronics Engineers

ER -

GUO YY, WONG ML, CAI ZH. A novel hybrid evolutionary algorithm for learning Bayesian networks from incomplete data. In Proceedings of the 2006 IEEE Congress on Evolutionary Computation, CEC 2006. Institute of Electrical and Electronics Engineers. 2006. p. 916-923 https://doi.org/10.1109/CEC.2006.1688409