Imprecise Gaussian Discriminant Classification - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue Pattern Recognition Année : 2019

Imprecise Gaussian Discriminant Classification

Résumé

Gaussian discriminant analysis is a popular classification model, that in the precise case can produce unreliable predictions in case of high uncertainty (scarce or noisy data set). While imprecise probability theory offer a nice theoretical framework to solve such issues, it has not been yet applied to Gaussian discriminant analysis. This work remedies this, by proposing a new Gaussian discriminant analysis based on robust Bayesian analysis and near-ignorance priors. The model delivers cautious predictions, in form of set-valued class, in case of limited or imperfect available information. Experiments show that including an imprecise component in the Gaussian discriminant analysis produce reasonably cautious predictions, and that set-valued predictions correspond to instances for which the precise model performs poorly.
Fichier principal
Vignette du fichier
imprecise_discriminant_analyse.pdf (1.12 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03386484 , version 1 (19-10-2021)

Identifiants

Citer

Yonatan Carlos Carranza-Alarcon, Sébastien Destercke. Imprecise Gaussian Discriminant Classification. Pattern Recognition, 2019, 112, pp.107739. ⟨10.1016/j.patcog.2020.107739⟩. ⟨hal-03386484⟩
30 Consultations
52 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More