Building and Exploiting Ontologies for an Automobile Project Memory

Joanna Golebiowska, Rose Dieng-Kuntz, Olivier Corby, Didier Mousseau

Abstract: This paper describes SAMOVAR (Systems Analysis of Modelling and Validation of Renault Automobiles), aiming at preserving and exploiting the memory of past projects in automobile design (in particular the memory of the problems encountered during a project) so as to exploit them in new projects. SAMOVAR relies on (1) the building of ontologies (in particular, thanks to the use of a linguistic tool on a textual corpus in order to enrich a core ontology in a semi-automatic way), (2) the �semantic� annotations of the descriptions of problems relatively to these ontologies, (3) the formalisation of the ontologies and annotations in RDF(S) so as to integrate in SAMOVAR the tool CORESE that enables an ontology-guided search in the base of the problem descriptions.