Contribution of belief functions to hidden markov models with an application to fault diagnosis.

Emmanuel Ramasso (1)

1 FEMTO-ST - Franche-Comté Électronique Mécanique, Thermique et Optique - Sciences et Technologies (UMR 6174)

Evidence-theoretic propagations of temporal belief functions are proposed to deal with possibly dependent observations and for partially supervised learning of HMM. Solutions are formulated in Transferable Belief Model framework and experiments concern a diagnosis problem.

Mots clés: Evidential hidden Markov models; state sequence recognition; partially supervised learning; system's health assessment.; system's health assessment

Domaines: Automatique / Robotique

Emmanuel Ramasso. Contribution of belief functions to hidden markov models with an application to fault diagnosis.. IEEE International Worshop on Machine Learning for Signal Processing, MLSP'09., Sep 2009, Grenoble, France. pp.1-6. ⟨hal-00414145⟩