Clustering Quality and Topology Preservation in Fast Learning SOMs

Salvatore Gaglio, Alfonso Urso, Giuseppe Di Fatta, Riccardo Rizzo, Antonino Fiannaca

Risultato della ricerca: Article

3 Citazioni (Scopus)

Abstract

The Self-Organizing Map (SOM) is a popular unsupervised neural network able to provide effective clustering and data visualization for data represented in multidimensional input spaces. In this paper, we describe Fast Learning SOM (FLSOM) which adopts a learning algorithm that improves the performance of the standard SOM with respect to the convergence time in the training phase. We show that FLSOM also improves the quality of the map by providing better clustering quality and topology preservation of multidimensional input data. Several tests have been carried out on different multidimensional datasets, which demonstrate better performances of the algorithm in comparison with the original SOM.
Lingua originaleEnglish
pagine (da-a)625-639
Numero di pagine15
RivistaNeural Network World
Volume19
Stato di pubblicazionePublished - 2009

Fingerprint

Self organizing maps
Cluster Analysis
Topology
Learning
Data visualization
Learning algorithms
Neural networks

All Science Journal Classification (ASJC) codes

  • Software
  • Neuroscience(all)
  • Hardware and Architecture
  • Artificial Intelligence

Cita questo

Gaglio, S., Urso, A., Di Fatta, G., Rizzo, R., & Fiannaca, A. (2009). Clustering Quality and Topology Preservation in Fast Learning SOMs. Neural Network World, 19, 625-639.

Clustering Quality and Topology Preservation in Fast Learning SOMs. / Gaglio, Salvatore; Urso, Alfonso; Di Fatta, Giuseppe; Rizzo, Riccardo; Fiannaca, Antonino.

In: Neural Network World, Vol. 19, 2009, pag. 625-639.

Risultato della ricerca: Article

Gaglio, S, Urso, A, Di Fatta, G, Rizzo, R & Fiannaca, A 2009, 'Clustering Quality and Topology Preservation in Fast Learning SOMs', Neural Network World, vol. 19, pagg. 625-639.
Gaglio S, Urso A, Di Fatta G, Rizzo R, Fiannaca A. Clustering Quality and Topology Preservation in Fast Learning SOMs. Neural Network World. 2009;19:625-639.
Gaglio, Salvatore ; Urso, Alfonso ; Di Fatta, Giuseppe ; Rizzo, Riccardo ; Fiannaca, Antonino. / Clustering Quality and Topology Preservation in Fast Learning SOMs. In: Neural Network World. 2009 ; Vol. 19. pagg. 625-639.
@article{03d44c2d2ee6440e947b3146f2148067,
title = "Clustering Quality and Topology Preservation in Fast Learning SOMs",
abstract = "The Self-Organizing Map (SOM) is a popular unsupervised neural network able to provide effective clustering and data visualization for data represented in multidimensional input spaces. In this paper, we describe Fast Learning SOM (FLSOM) which adopts a learning algorithm that improves the performance of the standard SOM with respect to the convergence time in the training phase. We show that FLSOM also improves the quality of the map by providing better clustering quality and topology preservation of multidimensional input data. Several tests have been carried out on different multidimensional datasets, which demonstrate better performances of the algorithm in comparison with the original SOM.",
keywords = "Clustering, FLSOM, SOM",
author = "Salvatore Gaglio and Alfonso Urso and {Di Fatta}, Giuseppe and Riccardo Rizzo and Antonino Fiannaca",
year = "2009",
language = "English",
volume = "19",
pages = "625--639",
journal = "Neural Network World",
issn = "1210-0552",
publisher = "Institute of Computer Science",

}

TY - JOUR

T1 - Clustering Quality and Topology Preservation in Fast Learning SOMs

AU - Gaglio, Salvatore

AU - Urso, Alfonso

AU - Di Fatta, Giuseppe

AU - Rizzo, Riccardo

AU - Fiannaca, Antonino

PY - 2009

Y1 - 2009

N2 - The Self-Organizing Map (SOM) is a popular unsupervised neural network able to provide effective clustering and data visualization for data represented in multidimensional input spaces. In this paper, we describe Fast Learning SOM (FLSOM) which adopts a learning algorithm that improves the performance of the standard SOM with respect to the convergence time in the training phase. We show that FLSOM also improves the quality of the map by providing better clustering quality and topology preservation of multidimensional input data. Several tests have been carried out on different multidimensional datasets, which demonstrate better performances of the algorithm in comparison with the original SOM.

AB - The Self-Organizing Map (SOM) is a popular unsupervised neural network able to provide effective clustering and data visualization for data represented in multidimensional input spaces. In this paper, we describe Fast Learning SOM (FLSOM) which adopts a learning algorithm that improves the performance of the standard SOM with respect to the convergence time in the training phase. We show that FLSOM also improves the quality of the map by providing better clustering quality and topology preservation of multidimensional input data. Several tests have been carried out on different multidimensional datasets, which demonstrate better performances of the algorithm in comparison with the original SOM.

KW - Clustering

KW - FLSOM

KW - SOM

UR - http://hdl.handle.net/10447/59641

M3 - Article

VL - 19

SP - 625

EP - 639

JO - Neural Network World

JF - Neural Network World

SN - 1210-0552

ER -