We would like to congratulate the authors on a well written article containing innovative ideas and a compelling application. The authors propose to incorporate subjective prior information on the clustering structure by defining a centred partition process. The proposed family of processes combines well known priors for partitions (specifically exchangeable partition probability functions, EPPFs) with a measure of discrepancy from an initial partition c_0, summarizing prior belief.

Contributed Discussion: Centered Partition Processes: Informative Priors for Clustering

Antoniano-Villalobos, Isadora
;
2021-01-01

Abstract

We would like to congratulate the authors on a well written article containing innovative ideas and a compelling application. The authors propose to incorporate subjective prior information on the clustering structure by defining a centred partition process. The proposed family of processes combines well known priors for partitions (specifically exchangeable partition probability functions, EPPFs) with a measure of discrepancy from an initial partition c_0, summarizing prior belief.
2021
16
File in questo prodotto:
File Dimensione Formato  
Comment_Paganin_etal.pdf

accesso aperto

Descrizione: Documento principale
Tipologia: Documento in Pre-print
Licenza: Creative commons
Dimensione 201.38 kB
Formato Adobe PDF
201.38 kB Adobe PDF Visualizza/Apri

I documenti in ARCA 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/10278/5004037
Citazioni
  • ???jsp.display-item.citation.pmc??? 2
  • Scopus 7
  • ???jsp.display-item.citation.isi??? 9
social impact