cbCPT: Knowledge Engineering Support for CPTs in Bayesian Networks

Abstract

Interacting with huge conditional probability tables in Bayesian belief networks makes it difficult for experts to create and employ probabilistic models. Researchers have investigated the use of graphical interfaces and knowledge engineering techniques to support experts’ interaction with complex BBNs. We propose a cased-based tool (cbCPT) especially designed to apply knowledge engineering principles to CPT navigation, elicitation, maintenance and evaluation. Using this approach experts and/or knowledge engineers can define particular scenarios to visualize and elicit their corresponding CPTs. Important cases defined by experts can be saved for further inspection and maintenance of CPTs. In addition, we report on a simple usability study in which participants walked through some particular cases.