Solutions obtained for a k-means algorithm depends heavily on the starting partition. In this article an aid is developed for implementing initialization methods particularly suited when all the cluster covariances matrices are presumed to be identical.
Mahalanobis metrics for k-means algorithms
TARSITANO, Agostino
2003-01-01
Abstract
Solutions obtained for a k-means algorithm depends heavily on the starting partition. In this article an aid is developed for implementing initialization methods particularly suited when all the cluster covariances matrices are presumed to be identical.File in questo prodotto:
Non ci sono file associati a questo prodotto.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.