A Parallel Sampling Method for Bayesian Networks

Yoshiki Kobari; Masaomi Kimura

doi:10.1109/ISCSIC.2017.11

A Parallel Sampling Method for Bayesian Networks

Yoshiki Kobari, Masaomi Kimura

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

2 Citations (Scopus)

Abstract

This paper describes the method to estimate probabilities on Bayesian belief networks (BNs). A BN has nodes showing random variables and shows cause and effect relationships among nodes as a graph. We calculate posterior probabilities, and then estimate the uncertain plural events. As one of the method for estimating probabilities in BN is stochastic sampling. The method has known as taking time to calculate as BN become larger and more complex. Therefore, this paper proposes a parallel sampling method on BN. Then a BN need efficiently dividing to generate samples in parallel, therefore we use community detection. The number of nodes, which have mutual dependence among nodes, is the least to use community detection. In addition, pipeline processing as generating samples, which have mutual dependence reduces waiting time causing by existing them.

Original language	English
Title of host publication	Proceedings - 2017 International Symposium on Computer Science and Intelligent Controls, ISCSIC 2017
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	113-117
Number of pages	5
Volume	2018-February
ISBN (Electronic)	9781538629413
DOIs	https://doi.org/10.1109/ISCSIC.2017.11
Publication status	Published - 2018 Feb 16
Event	1st International Symposium on Computer Science and Intelligent Controls, ISCSIC 2017 - Budapest, Hungary Duration: 2017 Oct 20 → 2017 Oct 22

Other

Other	1st International Symposium on Computer Science and Intelligent Controls, ISCSIC 2017
Country	Hungary
City	Budapest
Period	17/10/20 → 17/10/22

Keywords

Bayesian networks
dataminig
parallel sampling

ASJC Scopus subject areas

Computer Science Applications
Artificial Intelligence

Access to Document

10.1109/ISCSIC.2017.11

Fingerprint Dive into the research topics of 'A Parallel Sampling Method for Bayesian Networks'. Together they form a unique fingerprint.

Cite this

A Parallel Sampling Method for Bayesian Networks. / Kobari, Yoshiki; Kimura, Masaomi.

Proceedings - 2017 International Symposium on Computer Science and Intelligent Controls, ISCSIC 2017. Vol. 2018-February Institute of Electrical and Electronics Engineers Inc., 2018. p. 113-117 8294170.

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

Kobari, Y & Kimura, M 2018, A Parallel Sampling Method for Bayesian Networks. in Proceedings - 2017 International Symposium on Computer Science and Intelligent Controls, ISCSIC 2017. vol. 2018-February, 8294170, Institute of Electrical and Electronics Engineers Inc., pp. 113-117, 1st International Symposium on Computer Science and Intelligent Controls, ISCSIC 2017, Budapest, Hungary, 17/10/20. https://doi.org/10.1109/ISCSIC.2017.11

@inproceedings{ae49f23a0d7b4c3d8ba951157f030ee5,

title = "A Parallel Sampling Method for Bayesian Networks",

abstract = "This paper describes the method to estimate probabilities on Bayesian belief networks (BNs). A BN has nodes showing random variables and shows cause and effect relationships among nodes as a graph. We calculate posterior probabilities, and then estimate the uncertain plural events. As one of the method for estimating probabilities in BN is stochastic sampling. The method has known as taking time to calculate as BN become larger and more complex. Therefore, this paper proposes a parallel sampling method on BN. Then a BN need efficiently dividing to generate samples in parallel, therefore we use community detection. The number of nodes, which have mutual dependence among nodes, is the least to use community detection. In addition, pipeline processing as generating samples, which have mutual dependence reduces waiting time causing by existing them.",

keywords = "Bayesian networks, dataminig, parallel sampling",

author = "Yoshiki Kobari and Masaomi Kimura",

year = "2018",

month = feb,

day = "16",

doi = "10.1109/ISCSIC.2017.11",

language = "English",

volume = "2018-February",

pages = "113--117",

booktitle = "Proceedings - 2017 International Symposium on Computer Science and Intelligent Controls, ISCSIC 2017",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

note = "1st International Symposium on Computer Science and Intelligent Controls, ISCSIC 2017 ; Conference date: 20-10-2017 Through 22-10-2017",

}

TY - GEN

T1 - A Parallel Sampling Method for Bayesian Networks

AU - Kobari, Yoshiki

AU - Kimura, Masaomi

PY - 2018/2/16

Y1 - 2018/2/16

N2 - This paper describes the method to estimate probabilities on Bayesian belief networks (BNs). A BN has nodes showing random variables and shows cause and effect relationships among nodes as a graph. We calculate posterior probabilities, and then estimate the uncertain plural events. As one of the method for estimating probabilities in BN is stochastic sampling. The method has known as taking time to calculate as BN become larger and more complex. Therefore, this paper proposes a parallel sampling method on BN. Then a BN need efficiently dividing to generate samples in parallel, therefore we use community detection. The number of nodes, which have mutual dependence among nodes, is the least to use community detection. In addition, pipeline processing as generating samples, which have mutual dependence reduces waiting time causing by existing them.

AB - This paper describes the method to estimate probabilities on Bayesian belief networks (BNs). A BN has nodes showing random variables and shows cause and effect relationships among nodes as a graph. We calculate posterior probabilities, and then estimate the uncertain plural events. As one of the method for estimating probabilities in BN is stochastic sampling. The method has known as taking time to calculate as BN become larger and more complex. Therefore, this paper proposes a parallel sampling method on BN. Then a BN need efficiently dividing to generate samples in parallel, therefore we use community detection. The number of nodes, which have mutual dependence among nodes, is the least to use community detection. In addition, pipeline processing as generating samples, which have mutual dependence reduces waiting time causing by existing them.

KW - Bayesian networks

KW - dataminig

KW - parallel sampling

UR - http://www.scopus.com/inward/record.url?scp=85052383377&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85052383377&partnerID=8YFLogxK

U2 - 10.1109/ISCSIC.2017.11

DO - 10.1109/ISCSIC.2017.11

M3 - Conference contribution

AN - SCOPUS:85052383377

VL - 2018-February

SP - 113

EP - 117

BT - Proceedings - 2017 International Symposium on Computer Science and Intelligent Controls, ISCSIC 2017

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 1st International Symposium on Computer Science and Intelligent Controls, ISCSIC 2017

Y2 - 20 October 2017 through 22 October 2017

ER -