Construction of a Bayesian network as an extension of propositional logic

Takuto Enomoto, Masaomi Kimura

研究成果: Conference contribution

抄録

A Bayesian network is a probabilistic graphical model. Many conventional methods have been proposed for its construction. However, these methods often result in an incorrect Bayesian network structure. In this study, to correctly construct a Bayesian network, we extend the concept of propositional logic. We propose a methodology for constructing a Bayesian network with causal relationships that are extracted only if the antecedent states are true. In order to determine the logic to be used in constructing the Bayesian network, we propose the use of association rule mining such as the Apriori algorithm. We evaluate the proposed method by comparing its result with that of traditional method, such as Bayesian Dirichlet equivalent uniform (BDeu) score evaluation with a hill climbing algorithm, that shows that our method generates a network with more necessary arcs than that generated by the traditional method.

元の言語English
ホスト出版物のタイトルKDIR
出版者SciTePress
ページ211-217
ページ数7
1
ISBN(印刷物)9789897581588
出版物ステータスPublished - 2015
イベント7th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management, IC3K 2015 - Lisbon, Portugal
継続期間: 2015 11 122015 11 14

Other

Other7th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management, IC3K 2015
Portugal
Lisbon
期間15/11/1215/11/14

Fingerprint

Bayesian networks
Association rules

ASJC Scopus subject areas

  • Software

これを引用

Enomoto, T., & Kimura, M. (2015). Construction of a Bayesian network as an extension of propositional logic. : KDIR (巻 1, pp. 211-217). SciTePress.

Construction of a Bayesian network as an extension of propositional logic. / Enomoto, Takuto; Kimura, Masaomi.

KDIR. 巻 1 SciTePress, 2015. p. 211-217.

研究成果: Conference contribution

Enomoto, T & Kimura, M 2015, Construction of a Bayesian network as an extension of propositional logic. : KDIR. 巻. 1, SciTePress, pp. 211-217, 7th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management, IC3K 2015, Lisbon, Portugal, 15/11/12.
Enomoto T, Kimura M. Construction of a Bayesian network as an extension of propositional logic. : KDIR. 巻 1. SciTePress. 2015. p. 211-217
Enomoto, Takuto ; Kimura, Masaomi. / Construction of a Bayesian network as an extension of propositional logic. KDIR. 巻 1 SciTePress, 2015. pp. 211-217
@inproceedings{5bcb16b2a36748949931b4bee6fc824e,
title = "Construction of a Bayesian network as an extension of propositional logic",
abstract = "A Bayesian network is a probabilistic graphical model. Many conventional methods have been proposed for its construction. However, these methods often result in an incorrect Bayesian network structure. In this study, to correctly construct a Bayesian network, we extend the concept of propositional logic. We propose a methodology for constructing a Bayesian network with causal relationships that are extracted only if the antecedent states are true. In order to determine the logic to be used in constructing the Bayesian network, we propose the use of association rule mining such as the Apriori algorithm. We evaluate the proposed method by comparing its result with that of traditional method, such as Bayesian Dirichlet equivalent uniform (BDeu) score evaluation with a hill climbing algorithm, that shows that our method generates a network with more necessary arcs than that generated by the traditional method.",
keywords = "Association rule mining, Bayesian network, Propositional logic",
author = "Takuto Enomoto and Masaomi Kimura",
year = "2015",
language = "English",
isbn = "9789897581588",
volume = "1",
pages = "211--217",
booktitle = "KDIR",
publisher = "SciTePress",

}

TY - GEN

T1 - Construction of a Bayesian network as an extension of propositional logic

AU - Enomoto, Takuto

AU - Kimura, Masaomi

PY - 2015

Y1 - 2015

N2 - A Bayesian network is a probabilistic graphical model. Many conventional methods have been proposed for its construction. However, these methods often result in an incorrect Bayesian network structure. In this study, to correctly construct a Bayesian network, we extend the concept of propositional logic. We propose a methodology for constructing a Bayesian network with causal relationships that are extracted only if the antecedent states are true. In order to determine the logic to be used in constructing the Bayesian network, we propose the use of association rule mining such as the Apriori algorithm. We evaluate the proposed method by comparing its result with that of traditional method, such as Bayesian Dirichlet equivalent uniform (BDeu) score evaluation with a hill climbing algorithm, that shows that our method generates a network with more necessary arcs than that generated by the traditional method.

AB - A Bayesian network is a probabilistic graphical model. Many conventional methods have been proposed for its construction. However, these methods often result in an incorrect Bayesian network structure. In this study, to correctly construct a Bayesian network, we extend the concept of propositional logic. We propose a methodology for constructing a Bayesian network with causal relationships that are extracted only if the antecedent states are true. In order to determine the logic to be used in constructing the Bayesian network, we propose the use of association rule mining such as the Apriori algorithm. We evaluate the proposed method by comparing its result with that of traditional method, such as Bayesian Dirichlet equivalent uniform (BDeu) score evaluation with a hill climbing algorithm, that shows that our method generates a network with more necessary arcs than that generated by the traditional method.

KW - Association rule mining

KW - Bayesian network

KW - Propositional logic

UR - http://www.scopus.com/inward/record.url?scp=84960883615&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84960883615&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84960883615

SN - 9789897581588

VL - 1

SP - 211

EP - 217

BT - KDIR

PB - SciTePress

ER -