A two-level clustering for histogram data
Résumé
We present in this paper a clustering algorithm for histogram data based on a Self-Organising Map (SOM) learning that combine a dimension reduction by SOM and the clustering of the data in a reduced space in a certain number of homogeneous clusters. The number of cluster is not a priori fixed as parameter of the clustering algorithm but it is automatically found according to an estimation of local density of the data in the original space.
| Origine | Fichiers produits par l'(les) auteur(s) |
|---|---|
| Licence |