Learning various classes of models of lexicographic orderings

Booth, Richard; Chevaleyre, Yann; Lang, Jérôme; Mengin, Jérôme; Sombattheera, Chattrakul

Booth, Richard; Chevaleyre, Yann; Lang, Jérôme; Mengin, Jérôme; Sombattheera, Chattrakul (2009), Learning various classes of models of lexicographic orderings, Preference Learning (PL-09) ECML/PKDD-09 Workshop, Bled, SLOVENIA

View/Open

learning_various.PDF (192.6Kb)

learning_various_slides.PDF (101.2Kb)

Type

Communication / Conférence

Date

2009

Conference title

Preference Learning (PL-09) ECML/PKDD-09 Workshop

Conference city

Bled

Conference country

SLOVENIA

Metadata

Show full item record

Author(s)

Booth, Richard
Chevaleyre, Yann
Lang, Jérôme
Mengin, Jérôme
Sombattheera, Chattrakul

Abstract (EN)

We consider the problem of learning a user’s ordinal preferences onmultiattribute domains, assuming that the user’s preferences may be modelled asa kind of lexicographic ordering. We introduce a general graphical representationcalled LP-structures which captures various natural classes of such ordering inwhich both the order of importance between attributes and the local preferencesover each attributemay or may not be conditional on the values of other attributes.For each class we determine the Vapnik-Chernovenkis dimension, the communication complexity of learning preferences, and the complexity of identifying amodel in the class consistent with some given user-provided examples.

Subjects / Keywords

preferences; graphical representation; Complexity