Balancing the Elicitation Burden and the Richness of Expert Input When Quantifying Discrete Bayesian Networks.

Martine J Barons,Steven Mascaro,Anca M Hanea

doi:10.1111/risa.13772

Martine J Barons, Steven Mascaro + Show 1 more

Open Access

https://doi.org/10.1111/risa.13772

Copy DOI

Journal: Risk Analysis	Publication Date: Jun 19, 2021
Citations: 11	License type: CC BY 4.0

Affiliation: University of Warwick, University of Melbourne

Abstract

Structured expert judgment (SEJ) is a method for obtaining estimates of uncertain quantities from groups of experts in a structured way designed to minimize the pervasive cognitive frailties of unstructured approaches. When the number of quantities required is large, the burden on the groups of experts is heavy, and resource constraints may mean that eliciting all the quantities of interest is impossible. Partial elicitations can be complemented with imputation methods for the remaining, unelicited quantities. In the case where the quantities of interest are conditional probability distributions, the natural relationship between the quantities can be exploited to impute missing probabilities. Here we test the Bayesian intelligence interpolation method and its variations for Bayesian network conditional probability tables, called "InterBeta." We compare the various outputs of InterBeta on two cases where conditional probability tables were elicited from groups of experts. We show that interpolated values are in good agreement with experts' values and give guidance on how InterBeta could be used to good effect to reduce expert burden in SEJexercises.

Highlights

Modeling real-life problems often leads to highdimensional dependence or causal modeling of uncertain variables
Our experiments have not yet suggested a clear winner with respect to how well conditional probability table (CPT) fit their originals, and we focus here on our original interpolation method that is applied to the α and β
The aim is to show how accurate InterBeta is likely to be under various partial elicitation scenarios, and to provide guidance on which of the modes of operation might be optimal under different circumstances

Summary

Introduction

Modeling real-life problems often leads to highdimensional dependence or causal modeling of uncertain variables. Bayesian networks (BNs) are an established type of probabilistic graphical model that provides an elegant way of expressing the joint behavior of a large number of interrelated variables. BNs have been successfully used to represent uncertain knowledge, in a consistent probabilistic manner, in a variety of fields (Weber et al, 2010). They have the advantage that they are transparent. Mascaro and Hanea associated with each child node. These distributions serve as the quantitative information about the strength of the dependencies between the variables involved. The DAG with the conditional independence statements encoded by it, together with the (conditional) distributions, represents the joint distribution over the random variables denoted by the nodes of the graph

Objectives

Results

Conclusion