The present paper proposes a comparison of two fast algorithms for the classification of large binary data sets. There will be evaluated the incremental K-means and the light weight clustering (LWC) algorithms. In particular the focus is on the effective applicability on transactional data, having the procedures been developed for the analysis of data streams and microarrays. In addiction, it will be presented an improvement in the initialization phase of the incremental K-means algorithm.
A comparison of clustering methods for high dimensionaltransactional data
IODICE D'ENZA, Alfonso
2005-01-01
Abstract
The present paper proposes a comparison of two fast algorithms for the classification of large binary data sets. There will be evaluated the incremental K-means and the light weight clustering (LWC) algorithms. In particular the focus is on the effective applicability on transactional data, having the procedures been developed for the analysis of data streams and microarrays. In addiction, it will be presented an improvement in the initialization phase of the incremental K-means algorithm.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.
