This paper shows how typicality can be used to improve the case retrieval of a case-based reasoning (CBR) system, improving at the same time the global results of the CBR system. Typicality discriminates subclasses of a class in the domain ontology depending of how a subclass is a good example for its class. Our approach proposes to partition the subclasses of some classes into atypical, normal and typical subclasses in order to refine the domain ontology. The refined ontology allows a finergrained generalization of the query during the retrieval process. The benefits of this approach are presented according to an evaluation in the context of Taaable, a CBR system designed for the cooking domain.
CITATION STYLE
Gaillard, E., Lieber, J., & Nauer, E. (2015). Improving case retrieval using typicality. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 9343, pp. 165–180). Springer Verlag. https://doi.org/10.1007/978-3-319-24586-7_12
Mendeley helps you to discover research relevant for your work.