A latent logistic model to uncover overlapping clusters in networks

Pierre Latouche (1) , Etienne E. Birmelé (1) , Christophe Ambroise (1)

1 LSG - Laboratoire Statistique et Génome

It is now widely accepted that knowledge can be learnt from networks by clustering their vertices according to connection profiles. Many deterministic and probabilistic methods have been developed. Given a network, almost all them partition the vertices into disjoint clusters. However, recent studies have shown that these methods were too restrictive and that most of the existing networks contained overlapping clusters. To tackle this issue, we present in this paper a latent logistic model, that allows each vertex to belong to multiple clusters, as well as an efficient approximate inference procedure based on global and local variational techniques. We show the results that we obtained on a transcriptional regulatory network of yeast.

Domaines: Statistiques [math.ST]; Théorie [stat.TH]

Pierre Latouche, Etienne E. Birmelé, Christophe Ambroise. A latent logistic model to uncover overlapping clusters in networks. Atelier AGS (Apprentissage et Graphes pour les Systèmes complexes), May 2009, Hammamet, Tunisia. pp.3-8. ⟨hal-00629310⟩