A Model for Accountable Ordinal Sorting - Sorbonne Université Accéder directement au contenu
Communication Dans Un Congrès Année : 2017

A Model for Accountable Ordinal Sorting

Nicolas Maudet

Résumé

We address the problem of multicriteria ordinal sorting through the lens of accountability, i.e. the ability of a human decision-maker to own a recommendation made by the system. We put forward a number of model features that would favor the capability to support the recommendation with a convincing explanation. To account for that, we design a recommender system implementing and formalizing such features. This system outputs explanations under the form of specific argument schemes tailored to represent the specific rules of the model. At the end, we discuss possible and promising argumentative perspectives.
Fichier principal
Vignette du fichier
BelahceneEtAlIJCAI17.pdf (306.41 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01523779 , version 1 (15-10-2017)

Identifiants

  • HAL Id : hal-01523779 , version 1

Citer

Khaled Belahcene, Christophe Labreuche, Nicolas Maudet, Vincent Mousseau, Wassila Ouerdane. A Model for Accountable Ordinal Sorting. 26th International Joint Conference on Artificial Intelligence (IJCAI-2017), Aug 2017, Melbourne, Australia. ⟨hal-01523779⟩
190 Consultations
109 Téléchargements

Partager

Gmail Facebook X LinkedIn More