Title :
Contribution of belief functions to hidden markov models with an application to fault diagnosis
Author :
Ramasso, Emmanuel
Author_Institution :
Autom. Control & Micro-Mechatron. Syst. Dept., FEMTO-ST Inst., Besancon, France
Abstract :
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.
Keywords :
belief networks; fault diagnosis; hidden Markov models; learning (artificial intelligence); HMM; evidence-theoretic propagations; fault diagnosis; hidden Markov models; partially supervised learning; temporal belief functions; transferable belief model; Algorithm design and analysis; Automatic control; Bayesian methods; Data analysis; Fault diagnosis; Hidden Markov models; Inference algorithms; Possibility theory; Supervised learning; Uncertainty; Evidential Hidden Markov Models; Partially Supervised Learning; State Sequence Recognition; System´s Health Assessment;
Conference_Titel :
Machine Learning for Signal Processing, 2009. MLSP 2009. IEEE International Workshop on
Conference_Location :
Grenoble
Print_ISBN :
978-1-4244-4947-7
Electronic_ISBN :
978-1-4244-4948-4
DOI :
10.1109/MLSP.2009.5306209
