Clustering Sets of Objects Using Concepts-Objects Bipartite Graphs

Emmanuel Navarro (1) , Henri Prade (1) , Bruno Gaume (2)

1 IRIT-ADRIA - Argumentation, Décision, Raisonnement, Incertitude et Apprentissage
2 CLLE-ERSS - Cognition, Langues, Langage, Ergonomie

In this paper we deal with data stated under the form of abinary relation between objects and properties. We propose an approachfor clustering the objects and labeling them with characteristic subsetsof properties. The approach is based on a parallel between formal con-cept analysis and graph clustering. The problem is made tricky due tothe fact that generally there is no partitioning of the objects that can beassociated with a partitioning of properties. Indeed a relevant partitionof objects may exist, whereas it is not the case for properties. In order toobtain a conceptual clustering of the objects, we work with a bipartitegraph relating objects with formal concepts. Experiments on artificialbenchmarks and real examples show the effectiveness of the method,more particularly the fact that the results remain stable when an in-creasing number of properties are shared between objects of differentclusters.

Mots clés: graph clustering; formal concept analysis; bipartite graph

Domaines: Intelligence artificielle [cs.AI]

Emmanuel Navarro, Henri Prade, Bruno Gaume. Clustering Sets of Objects Using Concepts-Objects Bipartite Graphs. Sixth International Conference on Scalable Uncertainty Management (SUM 2012), Sep 2012, Marburg, Germany. pp.420-432, ⟨10.1007/978-3-642-33362-0_32⟩. ⟨hal-00992046⟩