SCHOOL OF COMPUTER SCIENCE AND SOFTWARE ENGINEERING
MONASH UNIVERSITY

TECHNICAL REPORT 2002/111



Knowledge Engineering Tools for Probability Elicitation

L R Hope, A E Nicholson and K B Korb

ABSTRACT

Unwieldy probability entry interfaces are the norm for Bayesian Network knowledge engineers.  This requires domain experts to provide unreasonably accurate probability estimates.  To solve these problems, we have developed two applications:  CPTable improves direct probability entry, and provides node customisation and sliding scale binary elicitation: Verbal Elicitor allows entry of probability values in ordinary English.  The domain expert selects a verbal cue such as "unlikely" or "almost certain."  The probabilities are then set manually or optimised to minimise probabilistic incoherency.