An Algorithm for Detecting Communities in Folksonomy Hypergraphs

Cécile Bothorel (1, 2) , Mohamed Bouklit (3)

1 Lab-STICC_TB_CID_DECIDE
2 LUSSI - Département Logique des Usages, Sciences sociales et Sciences de l'Information
3 FT - France Télécom, Division R&D, Lannion (France Télécom)

In this article, we are interested in social resource sharing systems such as Flickr, which use a lightweight knowledge representation called folksonomy. One of the fundamental questions asked by sociologists and actors involved in these online communities is to know whether a coherent tags categorization scheme emerges at global scale from folksonomy, though the users don’t share the same vocabulary. In order to satisfy their needs, we propose an algorithm to detect clusters in folksonomies hypergraphs by generalizing the Girvan and Newman’s clustering algorithm. We test our algorithm on a sample of an hypergragh of tag co-occurrence extracted from Flickr in September 2006, which gives promising results.

Domaines: Algorithme et structure de données [cs.DS]

Cécile Bothorel, Mohamed Bouklit. An Algorithm for Detecting Communities in Folksonomy Hypergraphs. I2CS 2011: 11th International Conference on Innovative Internet Community Services, Jun 2011, Berlin, Germany. pp.159 - 168. ⟨hal-00724991⟩