User menu

Accès à distance ? S'identifier sur le proxy UCLouvain

The Sum-over-Forests clustering

Primary tabs

Senelle, Mathieu [UCL] Saerens, Marco [UCL] Fouss, François [UCL]

This work introduces a novel way to identify dense regions in a graph based on a mode-seeking clustering technique, relying on the Sum-Over-Forests (SoF) density index (which can easily be computed in closed form through a simple matrix inversion) as a local density estimator. We first identify the modes of the SoF density in the graph. Then, the nodes of the graph are assigned to the cluster corresponding to the nearest mode, according to a new kernel, also based on the SoF framework. Experiments on artificial and real datasets show that the proposed index performs well in nodes clustering.

Document type Communication à un colloque (Conference Paper) – Présentation orale avec comité de sélection
Access type Accès restreint
Publication date 2014
Language Anglais
Conference "European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning", Bruges
Peer reviewed yes
Host document "Proceedings of the European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning"
Publication status Publié
Affiliations UCL - SSH/LouRIM - Louvain Research Institute in Management and Organizations
UCL - SST/ICTM/INGI - Pôle en ingénierie informatique
Links
Bibliographic reference Senelle, Mathieu ; Saerens, Marco ; Fouss, François. The Sum-over-Forests clustering.European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (Bruges). In: Proceedings of the European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, 2014
Permanent URL http://hdl.handle.net/2078.1/139927