Bayesian Network classifiers inferring workload from physiological features: compared performance

P. Besson , E. Dousset (1) , C. Bourdin (2) , L. Bringoux (2) , Tanguy Marqueste (3) , D. R. Mestre , J. L. Vercher

1 Laboratoire des Déterminants Physiologiques de l'Activité Physique
2 ISM - Institut des Sciences du Mouvement Etienne Jules Marey
3 CRMBM - Centre de résonance magnétique biologique et médicale

This paper presents an approach based on Bayesian Networks to estimate the workload of operators. The models take as inputs the entropy of different number of physiological features, as well as a cognitive feature (reaction time to a secondary task). They output the workload variation of subjects involved in successive tasks demanding different levels of cognitive resources. The performances of the classifiers are discussed in term of two criteria to be jointly optimized: the diversity, i.e. the ability of the model to perform on different subjects, and the accuracy, i.e., how close from the (subjectively estimated) workload level the model prediction is.

Domaines: Neurosciences

P. Besson, E. Dousset, C. Bourdin, L. Bringoux, Tanguy Marqueste, et al.. Bayesian Network classifiers inferring workload from physiological features: compared performance. 2012 IEEE INTELLIGENT VEHICLES SYMPOSIUM (IV), 2012, 345 E 47TH ST, NEW YORK, NY 10017 USA, Unknown Region. pp.282-287. ⟨hal-01436022⟩