Logo tutelle
Logo tutelle
Logo tutelle


Sur ce site

Sur le Web du CNRS

Accueil du site > Séminaires > Probabilités Statistiques et réseaux de neurones > Self-organizing maps and learning metrics

Vendredi 4 juin 2004, 10h-12h

Self-organizing maps and learning metrics

Samuel Kaski (University of Helsinki)

Résumé : I will start from introducing the basic principles and algorithms of the SOM and some of its variants. I will then discuss the use of SOMs for data exploration and text mining, and its relationships with some alternative methods. Learning metrics is a new principle for focusing data analysis on "relevant" aspects of data, instead of modeling both relevant and irrelevant parts. Metrics are learned from paired data. I will introduce the principle and discuss how to use with SOMs and other information visualization methods.

Dans la même rubrique :