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

Yuan Yuan GUO; Man Leung WONG; Zhi Hua CAI

doi:10.1109/CEC.2006.1688409

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

Yuan Yuan GUO
, Man Leung WONG
, Zhi Hua CAI

Department of Computing and Decision Sciences

Research output: Book Chapters | Papers in Conference Proceedings › Conference paper (refereed) › Research › peer-review

4 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 language	English
Title of host publication	Proceedings of the 2006 IEEE Congress on Evolutionary Computation, CEC 2006
Publisher	Institute of Electrical and Electronics Engineers
Pages	916-923
Number of pages	8
DOIs	https://doi.org/10.1109/CEC.2006.1688409
Publication status	Published - 1 Jan 2006

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

Access to Document

10.1109/CEC.2006.1688409

Cite this

@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 = jan,

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",

address = "United States",

}

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 Proceedings › Conference paper (refereed) › Research › peer-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 - https://www.scopus.com/pages/publications/34547317077

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 -

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

Abstract

Bibliographical note

Access to Document

Other files and links

Fingerprint

Cite this