|
Every picture tells a story: Visual Clustering in Relational Data| old_uid | 8849 |
|---|
| title | Every picture tells a story: Visual Clustering in Relational Data |
|---|
| start_date | 2010/06/07 |
|---|
| schedule | 09h30 |
|---|
| online | no |
|---|
| location_info | bat 25-26, salle 105 |
|---|
| summary | This 90 minute talk begins with definitions and examples of the three canonical problems of cluster analysis: tendency assessment, clustering, and cluster validity. The second part of this talk is a short history of visual approaches for these three problems. Part three defines and exemplifies the VAT (visual assessment of tendency) and improved iVAT approaches for building a reordered dissimilarity image from relational data that can be used to estimate cluster number and tendency. Part four describes an application of this algorithm to clustering ellipsoids in the context of wireless sensor networks. If time permits I will briefly discuss the sVAT algorithm (VAT for arbitrarily large square data), and coVAT (VAT for rectangular dissimilarity data. |
|---|
| responsibles | Bouchon-Meunier, Diaz, Gallinari |
|---|
| |
|