Computing diagnoses with higher posterior probability using Bayesian networks
DELCROIX ; PIECHOWIAK ; RODRIGUEZ
Type de document
COMMUNICATION AVEC ACTES INTERNATIONAL (ACTI)
Langue
anglais
Auteur
DELCROIX ; PIECHOWIAK ; RODRIGUEZ
Résumé / Abstract
This paper presents a method using Bayesian networks for diagnosing large devices. A network is described to model the device to diagnose. This model allows to take into account the structure of the device but also its behavior according to the state of its components. We propose an algorithm to find the best diagnoses, i.e. those with higher posterior probability, thanks to an approximation of this probability.
