CECM: Constrained Evidential C-Means algorithm

Violaine Antoine (1) , Benjamin Quost (1) , Marie-Hélène Masson (1) , Thierry Denoeux (1)

1 Heudiasyc - Heuristique et Diagnostic des Systèmes Complexes [Compiègne]

In clustering applications, prior knowledge about cluster membership is sometimes available. To integrate such auxiliary information, constraint-based (or semi-supervised) methods have been proposed in the hard or fuzzy clustering frameworks. This approach is extended to evidential clustering, in which the membership of objects to clusters is described by belief functions. A variant of the Evidential C-means (ECM) algorithm taking into account pairwise constraints is proposed. These constraints are translated into the belief function framework and integrated in the cost function. Experiments with synthetic and real data sets demonstrate the interest of the method. In particular, an application to medical image segmentation is presented.

Mots clés: Clustering; Semi-supervised learning; Pairwise constraints; Adaptive metric; Active learning; Belief functions; Dempster–Shafer theory; Evidence theory

Domaines: Intelligence artificielle [cs.AI]

Violaine Antoine, Benjamin Quost, Marie-Hélène Masson, Thierry Denoeux. CECM: Constrained Evidential C-Means algorithm. Computational Statistics and Data Analysis, 2012, 56 (4), pp.894-914. ⟨10.1016/j.csda.2010.09.021⟩. ⟨hal-00554310⟩