Parameter identification in Choquet Integral by the Kullback-Leibler diversgence on continuous densities with application to classification fusion.

Emmanuel Ramasso (1) , Sylvie Jullien (2)

1 FEMTO-ST - Franche-Comté Électronique Mécanique, Thermique et Optique - Sciences et Technologies (UMR 6174)
2 Dynamic3D

Classifier fusion is a means to increase accuracy and decision-making of classification systems by designing a set of basis classifiers and then combining their outputs. The combination is made up by non linear functional dependent on fuzzy measures called Choquet integral. It constitues a vast family of aggregation operators including minimum, maximum or weighted sum. The main issue before applying the Choquet integral is to identify the 2M − 2 parameters for M classifiers. We follow a previous work by Kojadinovic and one of the authors where the identification is performed using an informationtheoritic approach. The underlying probability densities are made smooth by fitting continuous parametric and then the Kullback-Leibler divergence is used to identify fuzzy measures. The proposed framework is applied on widely used datasets.

Mots clés: Classification.; Information fusion; Fuzzy measures; Relative entropy; Health assessment; Classification

Domaines: Automatique / Robotique

Emmanuel Ramasso, Sylvie Jullien. Parameter identification in Choquet Integral by the Kullback-Leibler diversgence on continuous densities with application to classification fusion.. European Society for Fuzzy Logic and Technology, EUSFLAT - LFA'11., Jul 2011, Aix-Les-Bains, France. pp.1-8. ⟨hal-00603967⟩