CBROnto: A Task/Method Ontology for CBR

May 2002

Proceedings of the Fifteenth International Florida Artificial Intelligence Research Society Conference (FLAIRS 2002)

Volume

Proceedings of the Fifteenth International Florida Artificial Intelligence Research Society Conference (FLAIRS 2002)

All Papers

Our approach to Case-Based Reasoning is to build integrated systems that combine case specific knowledge with models of general domain knowledge. In this paper we describe CBROnto, the CBR ontology we have developed, as a task/method ontology. CBROnto specifies a modelling framework to describe reusable CBR Problem Solving Methods based on the CBR tasks they solve and the knowledge requirements needed to apply them. We use Description Logics classification for two purposes. First, it is used to define a mapping mechanism to bridge the gap between methods and domains. And second, it is used to assess the applicability of a certain method to an application context, i.e. to check if the current application context satisfies the knowledge requirements of the method.