Evidential joint calibration of binary SVM classifiers
Résumé
In order to improve overall performance with respect to a classification problem, a path of research consists in using several classifiers and to fuse their outputs. To perform this fusion, some approaches merge the classifier outputs using a rule of combination. This requires that the outputs be made comparable beforehand, which is usually done thanks to a probabilistic calibration of each classifier. The fusion can also be performed by concatenating the classifier outputs into a vector and applying a joint probabilistic calibration to this vector. Recently, extensions of probabilistic calibration techniques of an individual classifier have been proposed using evidence theory, in order to better rep- resent the uncertainties inherent to the calibration process. In this paper, we adapt this latter idea to joint probabilistic calibration techniques, leading to evidential versions of joint calibration techniques. In addition, our proposal was tested on generated and real datasets and the results showed that it either outperforms or is comparable to state-of-the-art approaches.
Origine : Fichiers produits par l'(les) auteur(s)