A GA-based approach for parameter learning of discrete dynamic bayesian networks

Huange Wang; Xiaoguang Gao; C. P. Thompson

doi:10.1109/ICCMS.2010.126

A GA-based approach for parameter learning of discrete dynamic bayesian networks

Huange Wang, Xiaoguang Gao, C. P. Thompson

School of Electronics and Information

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Learning dynamic Bayesian networks (DBNs) is one of the current research focuses. In this article a GA-based approach is proposed for DBNs parameters learning from fully and partially observed data. The validity of the novel approach has been demonstrated by a detailedly described example, and the experimental results show that the proposed GA-based approach performs more accurately than the traditional EM algorithm.

Original language	English
Title of host publication	ICCMS 2010 - 2010 International Conference on Computer Modeling and Simulation
Pages	390-393
Number of pages	4
DOIs	https://doi.org/10.1109/ICCMS.2010.126
State	Published - 2010
Event	2010 International Conference on Computer Modeling and Simulation, ICCMS 2010 - Sanya, China Duration: 22 Jan 2010 → 24 Jan 2010

Publication series

Name	ICCMS 2010 - 2010 International Conference on Computer Modeling and Simulation
Volume	1

Conference

Conference	2010 International Conference on Computer Modeling and Simulation, ICCMS 2010
Country/Territory	China
City	Sanya
Period	22/01/10 → 24/01/10

Keywords

DBNs
EM algorithm
Genetic algorithm
Parameter learning

Access to Document

10.1109/ICCMS.2010.126

Cite this

@inproceedings{08a3670f2d8740d3a243a1b52b62b65c,

title = "A GA-based approach for parameter learning of discrete dynamic bayesian networks",

abstract = "Learning dynamic Bayesian networks (DBNs) is one of the current research focuses. In this article a GA-based approach is proposed for DBNs parameters learning from fully and partially observed data. The validity of the novel approach has been demonstrated by a detailedly described example, and the experimental results show that the proposed GA-based approach performs more accurately than the traditional EM algorithm.",

keywords = "DBNs, EM algorithm, Genetic algorithm, Parameter learning",

author = "Huange Wang and Xiaoguang Gao and Thompson, {C. P.}",

year = "2010",

doi = "10.1109/ICCMS.2010.126",

language = "英语",

isbn = "9780769539416",

series = "ICCMS 2010 - 2010 International Conference on Computer Modeling and Simulation",

pages = "390--393",

booktitle = "ICCMS 2010 - 2010 International Conference on Computer Modeling and Simulation",

note = "2010 International Conference on Computer Modeling and Simulation, ICCMS 2010 ; Conference date: 22-01-2010 Through 24-01-2010",

}

Wang, H, Gao, X & Thompson, CP 2010, A GA-based approach for parameter learning of discrete dynamic bayesian networks. in ICCMS 2010 - 2010 International Conference on Computer Modeling and Simulation., 5421364, ICCMS 2010 - 2010 International Conference on Computer Modeling and Simulation, vol. 1, pp. 390-393, 2010 International Conference on Computer Modeling and Simulation, ICCMS 2010, Sanya, China, 22/01/10. https://doi.org/10.1109/ICCMS.2010.126

A GA-based approach for parameter learning of discrete dynamic bayesian networks. / Wang, Huange; Gao, Xiaoguang; Thompson, C. P.
ICCMS 2010 - 2010 International Conference on Computer Modeling and Simulation. 2010. p. 390-393 5421364 (ICCMS 2010 - 2010 International Conference on Computer Modeling and Simulation; Vol. 1).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - A GA-based approach for parameter learning of discrete dynamic bayesian networks

AU - Wang, Huange

AU - Gao, Xiaoguang

AU - Thompson, C. P.

PY - 2010

Y1 - 2010

N2 - Learning dynamic Bayesian networks (DBNs) is one of the current research focuses. In this article a GA-based approach is proposed for DBNs parameters learning from fully and partially observed data. The validity of the novel approach has been demonstrated by a detailedly described example, and the experimental results show that the proposed GA-based approach performs more accurately than the traditional EM algorithm.

AB - Learning dynamic Bayesian networks (DBNs) is one of the current research focuses. In this article a GA-based approach is proposed for DBNs parameters learning from fully and partially observed data. The validity of the novel approach has been demonstrated by a detailedly described example, and the experimental results show that the proposed GA-based approach performs more accurately than the traditional EM algorithm.

KW - DBNs

KW - EM algorithm

KW - Genetic algorithm

KW - Parameter learning

UR - http://www.scopus.com/inward/record.url?scp=77951528137&partnerID=8YFLogxK

U2 - 10.1109/ICCMS.2010.126

DO - 10.1109/ICCMS.2010.126

M3 - 会议稿件

AN - SCOPUS:77951528137

SN - 9780769539416

T3 - ICCMS 2010 - 2010 International Conference on Computer Modeling and Simulation

SP - 390

EP - 393

BT - ICCMS 2010 - 2010 International Conference on Computer Modeling and Simulation

T2 - 2010 International Conference on Computer Modeling and Simulation, ICCMS 2010

Y2 - 22 January 2010 through 24 January 2010

ER -

A GA-based approach for parameter learning of discrete dynamic bayesian networks

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this