 |
|
|
|
| HAL : hal-00492745, version 2 |
| Fiche détaillée |  Récupérer au format |
|
|
| 16th European Signal Processing Conference EUSIPCO-2010, Aalborg, Denmark : France (2010) |
|
|
| Versions disponibles : | v1 (16-06-2010) | v2 (19-07-2010) |
|
|
|
|
| Clustering on Manifolds with Dual-Rooted Minimal Spanning Trees |
|
|
| Laurent Galluccio 1Olivier J.J. Michel 2 |
|
|
| (08/2010) |
|
|
|
| In this paper, we introduce a new distance computed from the construction of dual-rooted minimal spanning trees (MSTs). This distance extends Grikschat's approach, exhibits attractive properties and allows to account for both local and global neighborhood information. Furthermore, a function measuring the probability that a point belongs to a detected class is proposed. Some connections with diffusion maps are outlined. The dual-rooted tree-based distance (DRPT) allows us to construct a new affinity matrix for use in a spectral clustering algorithm, or leads to a new data analysis method. Results are presented on benchmark datasets. |
|
|
|
|
|
|
|
|
|
|
|
| 1 : | Laboratoire d'Informatique, Signaux, et Systèmes de Sophia-Antipolis (I3S) / Equipe SIGNAL |
|
|
|
Université de Nice Sophia Antipolis (UNS) – CNRS : UMR7271 |
|
|
| 2 : | Laboratoire d'Algorithmique Complexité et Logique (LACL) |
|
|
|
CNRS : FRE2673 – Université Paris XII - Paris Est Créteil Val-de-Marne |
|
|
|
|
|
|
|
|
|
| Domaine | : | Statistiques/Machine Learning |
|
|
| clustering – minimal spanning tree – distance measure |
|
|
|
| Liste des fichiers attachés à ce document : | |||||
|
|||||
|
|
|
| hal-00492745, version 2 | |
| http://hal.archives-ouvertes.fr/hal-00492745 | |
| oai:hal.archives-ouvertes.fr:hal-00492745 | |
| Contributeur : Laurent Galluccio | |
| Soumis le : Lundi 19 Juillet 2010, 16:30:01 | |
| Dernière modification le : Lundi 19 Juillet 2010, 17:02:50 | |
