Site-specific updating and aggregation of bayesian belief network models for multiple experts

Stiber, NA; Small, MJ; Pantazidou, M

dc.contributor.author	Stiber, NA	en
dc.contributor.author	Small, MJ	en
dc.contributor.author	Pantazidou, M	en
dc.date.accessioned	2014-03-01T11:44:38Z
dc.date.available	2014-03-01T11:44:38Z
dc.date.issued	2004	en
dc.identifier.issn	0272-4332	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/37055
dc.subject	Bayesian networks	en
dc.subject	Combining multiple expert beliefs	en
dc.subject	Environmental decision making	en
dc.subject	Expert systems	en
dc.subject	Reductive dechlorination	en
dc.subject.classification	Mathematics, Interdisciplinary Applications	en
dc.subject.classification	Social Sciences, Mathematical Methods	en
dc.subject.other	Chemicals	en
dc.subject.other	Contamination	en
dc.subject.other	Data reduction	en
dc.subject.other	Groundwater	en
dc.subject.other	Hazardous materials	en
dc.subject.other	Laws and legislation	en
dc.subject.other	Bayesian Belief Network (BBN)	en
dc.subject.other	Hazardous chemicals	en
dc.subject.other	Trichloroethene (TCE)	en
dc.subject.other	Risk assessment	en
dc.subject.other	ground water	en
dc.subject.other	Bayesian analysis	en
dc.subject.other	cleanup	en
dc.subject.other	groundwater pollution	en
dc.subject.other	methodology	en
dc.subject.other	Bayes theorem	en
dc.subject.other	dangerous goods	en
dc.subject.other	dechlorination	en
dc.subject.other	expert system	en
dc.subject.other	feasibility study	en
dc.subject.other	hazard assessment	en
dc.subject.other	mathematical analysis	en
dc.subject.other	review	en
dc.subject.other	risk assessment	en
dc.subject.other	water contamination	en
dc.subject.other	Algorithms	en
dc.subject.other	Artificial Intelligence	en
dc.subject.other	Bayes Theorem	en
dc.subject.other	Chlorine	en
dc.subject.other	Culture	en
dc.subject.other	Decision Making	en
dc.subject.other	Decision Making, Computer-Assisted	en
dc.subject.other	Decision Support Techniques	en
dc.subject.other	Demography	en
dc.subject.other	Expert Systems	en
dc.subject.other	Models, Statistical	en
dc.subject.other	Neural Networks (Computer)	en
dc.subject.other	Observer Variation	en
dc.subject.other	Probability	en
dc.subject.other	Risk	en
dc.subject.other	Trichloroethylene	en
dc.title	Site-specific updating and aggregation of bayesian belief network models for multiple experts	en
heal.type	other	en
heal.identifier.primary	10.1111/j.0272-4332.2004.00547.x	en
heal.identifier.secondary	http://dx.doi.org/10.1111/j.0272-4332.2004.00547.x	en
heal.language	English	en
heal.publicationDate	2004	en
heal.abstract	A method for combining multiple expert opinions that are encoded in a Bayesian Belief Network (BBN) model is presented and applied to a problem involving the cleanup of hazardous chemicals at a site with contaminated groundwater. The method uses Bayes Rule to update each expert model with the observed evidence, then uses it again to compute posterior probability weights for each model. The weights reflect the consistency of each model with the observed evidence, allowing the aggregate model to be tailored to the particular conditions observed in the site-specific application of the risk model. The Bayesian update is easy to implement, since the likelihood for the set of evidence (observations for selected nodes of the BBN model) is readily computed by sequential execution of the BBN model. The method is demonstrated using a simple pedagogical example and subsequently applied to a groundwater contamination problem using an expert-knowledge BBN model. The BBN model in this application predicts the probability that reductive dechlorination of the contaminant trichlorethene (TCE) is occurring at a site - a critical step in the demonstration of the feasibility of monitored natural attenuation for site cleanup - given information on 14 measurable antecedent and descendant conditions. The predictions for the BBN models for 21 experts are weighted and aggregated using examples of hypothetical and actual site data. The method allows more weight for those expert models that are more reflective of the site conditions, and is shown to yield an aggregate prediction that differs from that of simple model averaging in a potentially significant manner.	en
heal.publisher	BLACKWELL PUBLISHERS	en
heal.journalName	Risk Analysis	en
dc.identifier.doi	10.1111/j.0272-4332.2004.00547.x	en
dc.identifier.isi	ISI:000226235800011	en
dc.identifier.volume	24	en
dc.identifier.issue	6	en
dc.identifier.spage	1529	en
dc.identifier.epage	1538	en