Simplifying Case Retrieval in Case-based Reasoning using Ontologies

J.L. Castro, J.M. Sánchez, and J.M. Zurita (Spain)

Keywords

CaseBased Reasoning, Case Bases, Knowledge Representation, Ontologies

Abstract

Case-Based Reasoning (CBR) provides a methodology for reasoning (inferring conclusions) and learning with that reasoning process. One of the greatest limitations of these case-based reasoners is their close relationship with the kind of problem to be solved. This proximity to the problem stems from how information is represented in the case base. Since each problem is possibly specified in a different way, the reasoner must take this specification into account. This paper presents an alternative which considers a homogenous conception of the case base for all problems so that the involved CBR reasoning procedures may be standardized. In order to achieve this, in this first study, we shall focus on the conception of the case base with the aid of ontologies, so that cases may be recovered more effectively and efficiently in the initial phase of the CBR.

Important Links:



Go Back