Two-mode clustering consists in simultaneously partitioning modes (e.g., objects and variables) of an observed two-mode data matrix. A class of two-mode clustering algorithms in a fuzzy framework is proposed. Starting from the Double k-Means algorithm, different fuzzy proposals are addressed. The first one is the Fuzzy Double k-Means (FDkM) algorithm, providing two fuzzy partitions, one for each mode. A second proposal is the Fuzzy Double k-Means with polynomial fuzzifiers (FDkMpf) algorithm, a general method that includes the FDkM one as a particular case. Finally, a robust extension is introduced and analyzed by using the concept of noise cluster. The adequacy of the proposed algorithms is checked by means of a simulation and two real-case studies.

A class of two-mode clustering algorithms in a fuzzy setting / Ferraro, MARIA BRIGIDA; Giordani, Paolo; Vichi, Maurizio. - In: ECONOMETRICS AND STATISTICS. - ISSN 2452-3062. - 18:(2021), pp. 63-78. [10.1016/j.ecosta.2020.03.006]

A class of two-mode clustering algorithms in a fuzzy setting

Maria Brigida Ferraro
;
Paolo Giordani;Maurizio Vichi
2021

Abstract

Two-mode clustering consists in simultaneously partitioning modes (e.g., objects and variables) of an observed two-mode data matrix. A class of two-mode clustering algorithms in a fuzzy framework is proposed. Starting from the Double k-Means algorithm, different fuzzy proposals are addressed. The first one is the Fuzzy Double k-Means (FDkM) algorithm, providing two fuzzy partitions, one for each mode. A second proposal is the Fuzzy Double k-Means with polynomial fuzzifiers (FDkMpf) algorithm, a general method that includes the FDkM one as a particular case. Finally, a robust extension is introduced and analyzed by using the concept of noise cluster. The adequacy of the proposed algorithms is checked by means of a simulation and two real-case studies.
File allegati a questo prodotto
File Dimensione Formato  
Ferraro_two-model-clustering_2020.pdf

solo gestori archivio

Tipologia: Documento in Post-print (versione successiva alla peer review e accettata per la pubblicazione)
Licenza: Tutti i diritti riservati (All rights reserved)
Dimensione 684.1 kB
Formato Adobe PDF
684.1 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/11573/1390956
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 4
  • ???jsp.display-item.citation.isi??? 2
social impact