Structure learning of Bayesian networks using dual genetic algorithm

Lee Jaehun, Chung Wooyong, Euntai Kim

Research output: Contribution to journalArticle

13 Citations (Scopus)

Abstract

A new structure learning approach for Bayesian networks (BNs) based on dual genetic algorithm (DGA) is proposed in this paper. An individual of the population is represented as a dual chromosome composed of two chromosomes. The first chromosome represents the ordering among the BN nodes and the second represents the conditional dependencies among the ordered BN nodes. It is rigorously shown that there is no BN structure that cannot be encoded by the proposed dual genetic encoding and the proposed encoding explores the entire solution space of the BN structures. In contrast with existing GA-based structure learning methods, the proposed method learns not only the topology of the BN nodes, but also the ordering among the BN nodes, thereby, exploring the wider solution space of a given problem than the existing method. The dual genetic operators are closed in the set of the admissible individuals. The proposed method is applied to real-world and benchmark applications, while its effectiveness is demonstrated through computer simulation.

Original languageEnglish
Pages (from-to)32-43
Number of pages12
JournalIEICE Transactions on Information and Systems
VolumeE91-D
Issue number1
DOIs
Publication statusPublished - 2008 Jan 1

Fingerprint

Bayesian networks
Genetic algorithms
Chromosomes
Topology
Computer simulation

All Science Journal Classification (ASJC) codes

  • Software
  • Hardware and Architecture
  • Computer Vision and Pattern Recognition
  • Electrical and Electronic Engineering
  • Artificial Intelligence

Cite this

@article{6e7fcbaf71614ddca537d3704f6124cd,
title = "Structure learning of Bayesian networks using dual genetic algorithm",
abstract = "A new structure learning approach for Bayesian networks (BNs) based on dual genetic algorithm (DGA) is proposed in this paper. An individual of the population is represented as a dual chromosome composed of two chromosomes. The first chromosome represents the ordering among the BN nodes and the second represents the conditional dependencies among the ordered BN nodes. It is rigorously shown that there is no BN structure that cannot be encoded by the proposed dual genetic encoding and the proposed encoding explores the entire solution space of the BN structures. In contrast with existing GA-based structure learning methods, the proposed method learns not only the topology of the BN nodes, but also the ordering among the BN nodes, thereby, exploring the wider solution space of a given problem than the existing method. The dual genetic operators are closed in the set of the admissible individuals. The proposed method is applied to real-world and benchmark applications, while its effectiveness is demonstrated through computer simulation.",
author = "Lee Jaehun and Chung Wooyong and Euntai Kim",
year = "2008",
month = "1",
day = "1",
doi = "10.1093/ietisy/e91-d.1.32",
language = "English",
volume = "E91-D",
pages = "32--43",
journal = "IEICE Transactions on Information and Systems",
issn = "0916-8532",
publisher = "Maruzen Co., Ltd/Maruzen Kabushikikaisha",
number = "1",

}

Structure learning of Bayesian networks using dual genetic algorithm. / Jaehun, Lee; Wooyong, Chung; Kim, Euntai.

In: IEICE Transactions on Information and Systems, Vol. E91-D, No. 1, 01.01.2008, p. 32-43.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Structure learning of Bayesian networks using dual genetic algorithm

AU - Jaehun, Lee

AU - Wooyong, Chung

AU - Kim, Euntai

PY - 2008/1/1

Y1 - 2008/1/1

N2 - A new structure learning approach for Bayesian networks (BNs) based on dual genetic algorithm (DGA) is proposed in this paper. An individual of the population is represented as a dual chromosome composed of two chromosomes. The first chromosome represents the ordering among the BN nodes and the second represents the conditional dependencies among the ordered BN nodes. It is rigorously shown that there is no BN structure that cannot be encoded by the proposed dual genetic encoding and the proposed encoding explores the entire solution space of the BN structures. In contrast with existing GA-based structure learning methods, the proposed method learns not only the topology of the BN nodes, but also the ordering among the BN nodes, thereby, exploring the wider solution space of a given problem than the existing method. The dual genetic operators are closed in the set of the admissible individuals. The proposed method is applied to real-world and benchmark applications, while its effectiveness is demonstrated through computer simulation.

AB - A new structure learning approach for Bayesian networks (BNs) based on dual genetic algorithm (DGA) is proposed in this paper. An individual of the population is represented as a dual chromosome composed of two chromosomes. The first chromosome represents the ordering among the BN nodes and the second represents the conditional dependencies among the ordered BN nodes. It is rigorously shown that there is no BN structure that cannot be encoded by the proposed dual genetic encoding and the proposed encoding explores the entire solution space of the BN structures. In contrast with existing GA-based structure learning methods, the proposed method learns not only the topology of the BN nodes, but also the ordering among the BN nodes, thereby, exploring the wider solution space of a given problem than the existing method. The dual genetic operators are closed in the set of the admissible individuals. The proposed method is applied to real-world and benchmark applications, while its effectiveness is demonstrated through computer simulation.

UR - http://www.scopus.com/inward/record.url?scp=58149089351&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=58149089351&partnerID=8YFLogxK

U2 - 10.1093/ietisy/e91-d.1.32

DO - 10.1093/ietisy/e91-d.1.32

M3 - Article

VL - E91-D

SP - 32

EP - 43

JO - IEICE Transactions on Information and Systems

JF - IEICE Transactions on Information and Systems

SN - 0916-8532

IS - 1

ER -