Multi-Attribute Decision Making with Weighted Description Logics

We introduce a decision-theoretic framework based on Description Logics
\n(DLs), which can be used to encode and solve single stage multi-attribute decision problems. In particular, we consider the background knowledge as a DL
\nknowledge base where each attribute is represented by a concept, weighted by
\na utility value which is asserted by the user. This yields a compact representation of preferences over attributes. Moreover, we represent choices as knowledge
\nbase individuals, and induce a ranking via the aggregation of attributes that
\nthey satisfy. We discuss the benefits of the approach from a decision theory
\npoint of view. Furthermore, we introduce an implementation of the framework
\nas a Protégé plugin called uDecide. The plugin takes as input an ontology as
\nbackground knowledge, and returns the choices consistent with the user’s (the
\nknowledge base) preferences. We describe a use case with data from DBpedia.
\nWe also provide empirical results for its performance in the size of the ontology
\nusing the reasoner Konclude.