An Educational Tool for HighLevel Interaction with Bayesian Networks
Peter Haddawy Charles E. Kahn, Jr. Joel Jacobson
Department of EE & CS Department of Radiology University of Wisconsin--Milwaukee
Medical College of Wisconsin Milwaukee
We present an educational tool for bringing the in formation contained in a Bayesian network to the end user in an easily intelligible form. The banter shell is designed to tutor users in evaluation of hypotheses and selection of optimal diagnostic procedures. ban ter can be used with any Bayesian network contain ing nodes that can be classified into hypotheses, obser vations, and diagnostic procedures. We present algo rithms for determining optimal diagnostic procedures and for explanation generation.