Ordinal Learning with Vector Space Based Binary Predicates and Its Application to Tahitian Pearls’ Luster Automatic Assessment - Archive ouverte HAL Accéder directement au contenu
Chapitre D'ouvrage Année : 2018

Ordinal Learning with Vector Space Based Binary Predicates and Its Application to Tahitian Pearls’ Luster Automatic Assessment

G. Mondonneix
  • Fonction : Auteur
Sébastien Chabrier
Jean-Martial Mari
Alban Gabillon

Résumé

Algorithms designed to solve instance ranking problems are often a trade-off between classification and regression. We propose to solve instance ranking problems where the classes have fixed boundaries by observing that such cases can be reduced to object ranking problems. Object ranking implies determining a total order, which should imply in turn a computational cost exponential with the number of items to order. However, solving this problem in the feature space allows taking advantage of linearity, so as to ensure total order properties at no particular computational cost, in particular, without having to explicitly check for acyclicity. The proposed method is tested for classifying Tahitian pearls against their luster using photographs of commercial culture pearls ranked by experts of the profession and compared with previous support vector machine (SVM) multiclass classification. While the SVM approach had more than 20% of error (and more than 13% after feature selection), our method allows predicting the class of a pearl with less than 10% of error (and less than 8% after feature selection). Ordinal learning makes better use of implicit rank information and significantly (p < 10 −4) reduces classification error.
Fichier principal
Vignette du fichier
Mondonneix et al. - 2018 - Ordinal Learning with Vector Space Based Binary Predicates and Its Application to Tahitian Pearls’ Luster Aut.pdf (225.28 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-02144338 , version 1 (05-02-2021)

Identifiants

  • HAL Id : hal-02144338 , version 1

Citer

G. Mondonneix, Sébastien Chabrier, Jean-Martial Mari, Alban Gabillon. Ordinal Learning with Vector Space Based Binary Predicates and Its Application to Tahitian Pearls’ Luster Automatic Assessment: 8th International Conference, ICISP 2018, Cherbourg, France, July 2-4, 2018. Alamin Mansouri; Abderrahim El Moataz; Fathallah Nouboud; Driss Mammass. Image and Signal Processing, Springer, 2018, 978-3-319-94211-7. ⟨hal-02144338⟩

Collections

UPF
44 Consultations
61 Téléchargements

Partager

Gmail Facebook X LinkedIn More