Objects can be clustered in many different ways. As a matter of fact there are several cluster analysis methods that can produce different clusterings on the same dataset. Moreover, even when a single algorithm is used, different alternative clusterings can easily be generated, simply by changing the initial conditions of the algorithm. This work proposes a flexible criterion based on copula function for comparing two partitions (or clusterings) of the same dataset. This criterion also allows measuring the amount of information lost and gained in changing from cluster C to clustering C'.

Comparing clusterings by copula information based distance

Nai Ruscone, Marta
2017-01-01

Abstract

Objects can be clustered in many different ways. As a matter of fact there are several cluster analysis methods that can produce different clusterings on the same dataset. Moreover, even when a single algorithm is used, different alternative clusterings can easily be generated, simply by changing the initial conditions of the algorithm. This work proposes a flexible criterion based on copula function for comparing two partitions (or clusterings) of the same dataset. This criterion also allows measuring the amount of information lost and gained in changing from cluster C to clustering C'.
2017
978-88-99459-71-0
File in questo prodotto:
File Dimensione Formato  
5184.pdf

accesso chiuso

Tipologia: Documento in versione editoriale
Dimensione 308.06 kB
Formato Adobe PDF
308.06 kB Adobe PDF   Visualizza/Apri   Richiedi una copia

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11567/1013391
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact