Data Driven Concept Refinement to Support Avionics Maintenance

Luis Palacios Medinacelli (1, 2) , Yue Ma (1) , Gaëlle Lortal (2) , Claire Laudy (2) , Chantal Reynaud (1) , Luis Palacios (1)

1 LaHDAK - LRI - Données et Connaissances Massives et Hétérogènes (LRI)
2 Thales Research and Technology [Palaiseau]

Description Logic Ontologies are one of the most important knowledge representation formalisms nowadays which, broadly speaking, consist of classes of objects and their relations. Given a set of objects as samples and a class expression describing them, we present ongoing work that formalizes which properties of these objects are the most relevant for the given class expression to capture them. Moreover , we provide guidance on how to refine the given expression to better describe the set of objects. The approach is used to characterize test results that lead to a specific maintenance corrective action, and in this paper is illustrated to define sub-classes of aviation reports related to specific aircraft equipment.

Domaines: Intelligence artificielle [cs.AI]

Luis Palacios Medinacelli, Yue Ma, Gaëlle Lortal, Claire Laudy, Chantal Reynaud, et al.. Data Driven Concept Refinement to Support Avionics Maintenance. Proceedings of the IJCAI Workshop on Semantic Machine Learning , Aug 2017, Melbourne, Australia. ⟨hal-01632675⟩