Skip to Main Content (Press Enter)

Logo UNICH
  • ×
  • Home
  • Degrees
  • Courses
  • Jobs
  • People
  • Outputs
  • Organizations
  • Third Mission
  • Projects
  • Expertise & Skills

UNI-FIND
Logo UNICH

|

UNI-FIND

unich.it
  • ×
  • Home
  • Degrees
  • Courses
  • Jobs
  • People
  • Outputs
  • Organizations
  • Third Mission
  • Projects
  • Expertise & Skills
  1. Outputs

A new functional clustering method: the functional clustering and dimensional reduction model

Chapter
Publication Date:
2022
abstract:
In this work a new general procedure to cluster functional observations
in a subspace of reduced dimension is proposed. In particular the method simultaneously performs cluster analysis and dimension reduction of functional data. The
Functional Clustering and Dimension Reduction (FCDR) is obtained as the combination of two objective functions: the Functional Principal Component Analysis
(FPCA) and the Functional Factorial K-Means (FFKM). The advantage of this approach consists in the optimization of a single global objective function which, by
means of the selection of a tuning parameter, incorporates several techniques varying from the FPCA to FFKM including intermediate cases of clustering and dimension reduction.
Abstract In questo lavoro viene proposta una nuova procedura generale per raggruppare dati funzionali in un sottospazio di dimensioni ridotte. In particolare, il
metodo esegue simultaneamente la cluster analysis e la riduzione della dimensionalita dei dati funzionali. Il metodo Functional Clustering and Dimension Reduc- `
tion (FCDR) e ottenuto come la combinazione di due funzioni obiettivo: la Func- `
tional Principal Component Analysis (FPCA) ed il Functional Factorial K-Means
(FFKM). Il vantaggio di questo approccio consiste nell’ottimizzare una sola funzione obiettivo globale che, attraverso la selezione di un parametro di controllo,
include diverse tecniche che vanno dalla FPCA al FFKM includendo casi intermedi
di raggruppamento e riduzione dei dati.
Iris type:
2.1 Contributo in volume (Capitolo o Saggio)
Keywords:
Cluster Analysis, Functional Data, Dimension Reduction
List of contributors:
Evangelista, Adelia; Gattone, Stefano Antonio
Authors of the University:
EVANGELISTA Adelia
GATTONE Stefano Antonio
Handle:
https://ricerca.unich.it/handle/11564/804612
Full Text:
https://ricerca.unich.it//retrieve/handle/11564/804612/371706/SIS2022_2.zip
Book title:
SIS 2022 Book of Short Papers
  • Overview

Overview

URL

https://it.pearson.com/docenti/universita/partnership/sis.html
  • Use of cookies

Powered by VIVO | Designed by Cineca | 26.5.2.0