Neural and statistical classifiers. Can such approaches be complementary? - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2000

Neural and statistical classifiers. Can such approaches be complementary?

Christian Barat
Etienne Colle
  • Fonction : Auteur
  • PersonId : 855930
Sylvie Lelandais
  • Fonction : Auteur
  • PersonId : 840428

Résumé

Neural networks are efficient in certain pattern recognition sub-problems, especially in feature extraction and classification. In many cases neural and statistical techniques are seen as alternatives. Our aim is to verify if these approaches can give complementary responses in order to consider the implementation of fusion methods. The comparison is applied to three examples belonging to mobile robot localization: (i) laser range finder modeling, (ii) feature extraction from ultrasonic range finder data and (iii) localization by a stereoscopic camera. In each case the solution of the problem is based partly on a classifier. The paper compares the performances of a multilayer perceptron (MLP) known as an efficient classifier and three statistical methods-quadratic discriminant analysis (QDA), linear discriminant analysis (LDA) and Bayesian. The performances of the classifier are estimated by classical criteria such as success and misclassification percentages and the study is completed by a sharp analysis where the method results are crossed two by two to evaluate the success percentage of a method applied to the misclassified set of another one. Experiments show the set of patterns misclassified by the different classifiers does not completely overlap.
Fichier non déposé

Dates et versions

hal-01636544 , version 1 (16-11-2017)

Identifiants

Citer

Christian Barat, Humberto Loaiza, Etienne Colle, Sylvie Lelandais. Neural and statistical classifiers. Can such approaches be complementary?. IEEE Instrumentation and Measurement Technology Conference, May 2000, Baltimore, MD, United States. ⟨10.1109/IMTC.2000.848720⟩. ⟨hal-01636544⟩

Collections

CEA UNIV-EVRY
91 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More