Integrated modular Bayesian networks with selective inference for context-aware decision making

Seung Hyun Lee, Kyon Mo Yang, Sung Bae Cho

Research output: Contribution to journalArticle

10 Citations (Scopus)

Abstract

As many devices equipped with various sensors have recently proliferated, the fusion methods of various information and data from different sources have been studied. Bayesian network is one of the popular methods that solve this problem to cope with the uncertainty and imprecision. However, because a monolithic Bayesian network has high computational and design complexities, it is hard to apply to realistic problems. In this paper, we propose a modular Bayesian network system to extract context information by cooperative inference of multiple modules, which guarantees reliable inference compared to a monolithic Bayesian network without losing its strength like the easy management of knowledge and scalability. The proposed method preserves inter-modular dependencies by virtual linking and has lower computational complexity in complicated environments. The inter-modular d-separation controls local information to be delivered only to relevant modules. We verify that the proposed modular Bayesian network is enough to keep inter-modular causalities in a time-saving manner. This paper shows a possibility that a context-aware system would be easily constructed by mashing up Bayesian network fractions independently designed or leaned in different domains.

Original languageEnglish
Pages (from-to)38-46
Number of pages9
JournalNeurocomputing
Volume163
DOIs
Publication statusPublished - 2015 Sep 2

Fingerprint

Bayesian networks
Decision Making
Decision making
Knowledge Management
Information Storage and Retrieval
Causality
Uncertainty
Equipment and Supplies
Scalability
Computational complexity
Fusion reactions
Sensors

All Science Journal Classification (ASJC) codes

  • Computer Science Applications
  • Cognitive Neuroscience
  • Artificial Intelligence

Cite this

@article{fba76a299a4e4e53971ebbf4f2dd6b77,
title = "Integrated modular Bayesian networks with selective inference for context-aware decision making",
abstract = "As many devices equipped with various sensors have recently proliferated, the fusion methods of various information and data from different sources have been studied. Bayesian network is one of the popular methods that solve this problem to cope with the uncertainty and imprecision. However, because a monolithic Bayesian network has high computational and design complexities, it is hard to apply to realistic problems. In this paper, we propose a modular Bayesian network system to extract context information by cooperative inference of multiple modules, which guarantees reliable inference compared to a monolithic Bayesian network without losing its strength like the easy management of knowledge and scalability. The proposed method preserves inter-modular dependencies by virtual linking and has lower computational complexity in complicated environments. The inter-modular d-separation controls local information to be delivered only to relevant modules. We verify that the proposed modular Bayesian network is enough to keep inter-modular causalities in a time-saving manner. This paper shows a possibility that a context-aware system would be easily constructed by mashing up Bayesian network fractions independently designed or leaned in different domains.",
author = "Lee, {Seung Hyun} and Yang, {Kyon Mo} and Cho, {Sung Bae}",
year = "2015",
month = "9",
day = "2",
doi = "10.1016/j.neucom.2014.08.089",
language = "English",
volume = "163",
pages = "38--46",
journal = "Review of Economic Dynamics",
issn = "1094-2025",
publisher = "Academic Press Inc.",

}

Integrated modular Bayesian networks with selective inference for context-aware decision making. / Lee, Seung Hyun; Yang, Kyon Mo; Cho, Sung Bae.

In: Neurocomputing, Vol. 163, 02.09.2015, p. 38-46.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Integrated modular Bayesian networks with selective inference for context-aware decision making

AU - Lee, Seung Hyun

AU - Yang, Kyon Mo

AU - Cho, Sung Bae

PY - 2015/9/2

Y1 - 2015/9/2

N2 - As many devices equipped with various sensors have recently proliferated, the fusion methods of various information and data from different sources have been studied. Bayesian network is one of the popular methods that solve this problem to cope with the uncertainty and imprecision. However, because a monolithic Bayesian network has high computational and design complexities, it is hard to apply to realistic problems. In this paper, we propose a modular Bayesian network system to extract context information by cooperative inference of multiple modules, which guarantees reliable inference compared to a monolithic Bayesian network without losing its strength like the easy management of knowledge and scalability. The proposed method preserves inter-modular dependencies by virtual linking and has lower computational complexity in complicated environments. The inter-modular d-separation controls local information to be delivered only to relevant modules. We verify that the proposed modular Bayesian network is enough to keep inter-modular causalities in a time-saving manner. This paper shows a possibility that a context-aware system would be easily constructed by mashing up Bayesian network fractions independently designed or leaned in different domains.

AB - As many devices equipped with various sensors have recently proliferated, the fusion methods of various information and data from different sources have been studied. Bayesian network is one of the popular methods that solve this problem to cope with the uncertainty and imprecision. However, because a monolithic Bayesian network has high computational and design complexities, it is hard to apply to realistic problems. In this paper, we propose a modular Bayesian network system to extract context information by cooperative inference of multiple modules, which guarantees reliable inference compared to a monolithic Bayesian network without losing its strength like the easy management of knowledge and scalability. The proposed method preserves inter-modular dependencies by virtual linking and has lower computational complexity in complicated environments. The inter-modular d-separation controls local information to be delivered only to relevant modules. We verify that the proposed modular Bayesian network is enough to keep inter-modular causalities in a time-saving manner. This paper shows a possibility that a context-aware system would be easily constructed by mashing up Bayesian network fractions independently designed or leaned in different domains.

UR - http://www.scopus.com/inward/record.url?scp=84930271527&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84930271527&partnerID=8YFLogxK

U2 - 10.1016/j.neucom.2014.08.089

DO - 10.1016/j.neucom.2014.08.089

M3 - Article

AN - SCOPUS:84930271527

VL - 163

SP - 38

EP - 46

JO - Review of Economic Dynamics

JF - Review of Economic Dynamics

SN - 1094-2025

ER -