Approaches to bayesian network structure elicitation

K. Zwirglmaier, D. Straub

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

3 Scopus citations

Abstract

Bayesian networks (BN) are increasingly popular as a modeling tool for engineering risk and reliability analysis. BNs consist of two parts: the qualitative dependence structure represented by a directed acyclic graph (DAG), and the quantification of local dependences through conditional probability distributions (CPDs). The derivation of the general dependence structure requires a profound understanding of the problem as well as an understanding of the modeling tool. Once the general structure is developed, the probability expert knows what CPDs need to be elicited through data, expert estimates or a combination of both.

Original languageEnglish
Title of host publicationRisk, Reliability and Safety
Subtitle of host publicationInnovating Theory and Practice - Proceedings of the 26th European Safety and Reliability Conference, ESREL 2016
EditorsLesley Walls, Matthew Revie, Tim Bedford
PublisherCRC Press/Balkema
Pages56
Number of pages1
ISBN (Print)9781138029972
StatePublished - 2017
Event26th European Safety and Reliability Conference, ESREL 2016 - Glasgow, United Kingdom
Duration: 25 Sep 201629 Sep 2016

Publication series

NameRisk, Reliability and Safety: Innovating Theory and Practice - Proceedings of the 26th European Safety and Reliability Conference, ESREL 2016

Conference

Conference26th European Safety and Reliability Conference, ESREL 2016
Country/TerritoryUnited Kingdom
CityGlasgow
Period25/09/1629/09/16

Fingerprint

Dive into the research topics of 'Approaches to bayesian network structure elicitation'. Together they form a unique fingerprint.

Cite this