Possibilistic Preference Networks and Lexicographic Preference Trees - A Comparison
Résumé
The paper compares two graphical approaches proposed for the qualitative modeling of preferences: π-pref nets and LP-trees. The former uses the graphical setting of possibilistic networks for completing partial specifications of user preferences, while the latter, which is based on lexicographic ordering, appears to offer a convenient framework for learning preferences. The π-pref network representation appears to be more flexible, even if the addition of very specific constraints allows us to recover the total order of the LP-trees.
Domaines
| Origine | Fichiers produits par l'(les) auteur(s) |
|---|---|
| Licence |