Dynamic Probabilistic CP-nets

In this paper we present a two-fold generalization of conditional preference networks (CP-nets) that incorporates uncertainty. CP-nets are a for- mal tool to model qualitative conditional statements (cp-statements) about preferences over a set of objects. They are inherently static structures, both in their ability to capture dependencies between objects and in their expression of preferences over features of a particular object. Moreover, CP-nets do not provide the ability to express uncertainty over the preference statements. We present and study a generalization of CP-nets which supports changes and allows for encoding uncertainty, expressed in probabilistic terms, over the structure of the dependency links and over the individual preference relations