Space-time FPCA Algorithm for clustering of multidimensional curves.

Risultato della ricerca: Paper

Abstract

In this paper we focus on finding clusters of multidimensional curves with spatio-temporal structure, applying a variant of a k-means algorithm based on the principal component rotation of data. The main advantage of this approach is to combine the clustering functional analysis of the multidimensional data, with smoothing methods based on generalized additive models, that cope with both the spatial and the temporal variability, and with functional principal components that takes into account the dependency between the curves.
Lingua originaleEnglish
Stato di pubblicazionePublished - 2016

Fingerprint

Functional analysis

Cita questo

@conference{d560fecda6ed4ede89f28fa63c8297f3,
title = "Space-time FPCA Algorithm for clustering of multidimensional curves.",
abstract = "In this paper we focus on finding clusters of multidimensional curves with spatio-temporal structure, applying a variant of a k-means algorithm based on the principal component rotation of data. The main advantage of this approach is to combine the clustering functional analysis of the multidimensional data, with smoothing methods based on generalized additive models, that cope with both the spatial and the temporal variability, and with functional principal components that takes into account the dependency between the curves.",
author = "Marcello Chiodi and {Di Salvo}, Francesca and Giada Adelfio",
year = "2016",
language = "English",

}

TY - CONF

T1 - Space-time FPCA Algorithm for clustering of multidimensional curves.

AU - Chiodi, Marcello

AU - Di Salvo, Francesca

AU - Adelfio, Giada

PY - 2016

Y1 - 2016

N2 - In this paper we focus on finding clusters of multidimensional curves with spatio-temporal structure, applying a variant of a k-means algorithm based on the principal component rotation of data. The main advantage of this approach is to combine the clustering functional analysis of the multidimensional data, with smoothing methods based on generalized additive models, that cope with both the spatial and the temporal variability, and with functional principal components that takes into account the dependency between the curves.

AB - In this paper we focus on finding clusters of multidimensional curves with spatio-temporal structure, applying a variant of a k-means algorithm based on the principal component rotation of data. The main advantage of this approach is to combine the clustering functional analysis of the multidimensional data, with smoothing methods based on generalized additive models, that cope with both the spatial and the temporal variability, and with functional principal components that takes into account the dependency between the curves.

UR - http://hdl.handle.net/10447/180210

M3 - Paper

ER -