A review of relationships between possibility and probability representations of uncertainty in measurement

Abstract : The main advances regarding the deep connections between probability and possibility measurement uncertainty representation (but not the propagation) over the last decade are reviewed. They concern: the definition of a possibility distribution equivalent to a probability one from its whole set of dispersion intervals for all the probability levels, the bridges with the conventional dispersion parameters, the representation of a partial probability knowledge thanks to a maximum specificity principle better than the maximum entropy principle and also related to probability inequalities. The use of a possibility representation for common measurement situations such as the description of measurement results, measurand estimation and the expression of a priori uncertainty information are illustrated, and then discussed in view of their use in further processing (propagation, fuzzy inference systems). The conclusion highlights the interests of the possibility approach and points out some remaining issues.
Type de document :
Article dans une revue
IEEE Transactions on Instrumentation and Measurement, Institute of Electrical and Electronics Engineers, 2013, 62 (3), pp.622-632. 〈10.1109/TIM.2012.2218057〉
Liste complète des métadonnées

http://hal.univ-smb.fr/hal-00829150
Contributeur : Gilles Mauris <>
Soumis le : dimanche 2 juin 2013 - 10:50:53
Dernière modification le : mercredi 10 janvier 2018 - 09:43:53

Identifiants

Collections

Citation

Gilles Mauris. A review of relationships between possibility and probability representations of uncertainty in measurement. IEEE Transactions on Instrumentation and Measurement, Institute of Electrical and Electronics Engineers, 2013, 62 (3), pp.622-632. 〈10.1109/TIM.2012.2218057〉. 〈hal-00829150〉

Partager

Métriques

Consultations de la notice

154