Integrating ontologies and case-based reasoning for the development of knowledge-intensive intelligent systems

Case-Based Reasoning (CBR) allows emulating the human inference of solutions to problems profiting from previous experience. The integration of CBR with ontologies, structured organization of semantic knowledge, has been in the attention for some time, aiming to create powerful knowledge-intensive systems capable of proposing appropriate solutions to problems. This entails having collected an appropriate number of previous cases as well as having established a suitable ontology for the application domain. This paper focuses on the integration of CBR and ontologies to support the case representation, case base storage, and semantic similarity estimation. Different alternatives for such integration are explored and the approach has been tested in the creation of a Decision-Support System for the design of predictive maintenance systems. This work opens the window to significantly improve the capabilities of a CBR system by using the knowledge materialization and the reasoning features of a specific domain ontology.