Thumbnail

The Complexity of Learning Separable ceteris paribus Preferences

Lang, Jérôme; Mengin, Jérôme (2009), The Complexity of Learning Separable ceteris paribus Preferences, IJCAI-09, 21st International Joint Conference on Artificial Intelligence, AAAI Press / IJCAI : Palo Alto (USA), p. 848-853

View/Open
Type
Communication / Conférence
Date
2009
Conference country
UNITED STATES
Book title
IJCAI-09, 21st International Joint Conference on Artificial Intelligence
Publisher
AAAI Press / IJCAI
Published in
Palo Alto (USA)
Pages
848-853
Metadata
Show full item record
Author(s)
Lang, Jérôme
Mengin, Jérôme
Abstract (EN)
We address the problem of learning preference relations on multi-attribute (or combinatorial) domains.We do so by making a very simple hypothesis aboutthe dependence structure between attributes thatthe preference relation enjoys, namely separability(no preferential dependencies between attributes).Given a set of examples consisting of comparisonsbetween alternatives, we want to output a separableCP-net, consisting of local preferences on each ofthe attributes, that fits the examples. We considerthree forms of compatibility between a CP-net anda set of examples, and for each of them we give useful characterizations as well as complexity results.
Subjects / Keywords
Preference relations