Heterogeneity in ontological CBR systems

Amjad Abou Assali (1) , Dominique Lenne (1) , Bruno Debray (2)

1 Heudiasyc - Heuristique et Diagnostic des Systèmes Complexes [Compiègne]
2 INERIS - Institut National de l'Environnement Industriel et des Risques

We present in this chapter our platform COBRA, an Ontology-based Case-based Reasoning (CBR) platform. We work currently on the diagnosis of the failures of gas sensors installed at industrial sites. COBRA allows to author and reuse past experiences in order to diagnose new failure situations. However, it can be used, in general, to develop CBR systems for classification tasks. COBRA is based on a knowledge base that integrates domain knowledge along with cases in an ontological structure, which enhances its semantic reasoning capacities. Users can describe their cases using instances from the knowledge base. The resulting case base is heterogeneous where cases do not always share the same attributes. We focus, in this chapter, on the problems of heterogeneity, and we present our case alignment approach allowing to treat this heterogeneity.

Domaines: Sciences de l'ingénieur [physics]

Amjad Abou Assali, Dominique Lenne, Bruno Debray. Heterogeneity in ontological CBR systems. Montani, S., Jain, L.C. Successful case-based reasoning applications, Springer, pp.97-116, 2011, Studies in computational intelligence, 978-3-642-14078-5. ⟨10.1007/978-3-642-14078-5_5⟩. ⟨ineris-00969427⟩