Cluster analysis is aimed at classifying elements into categories on the basis of their similarity. Its applications range from astronomy to bioinformatics, bibliometrics, and pattern recognition.We propose an approach based on the idea that cluster centers are characterized by a higher density than their neighbors and by a relatively large distance from points with higher densities. This idea forms the basis of a clustering procedure in which the number of clusters arises intuitively, outliers are automatically spotted and excluded fromthe analysis, and clusters are recognized regardless of their shape and of the dimensionality of the space inwhich they are embedded.We demonstrate the power of the algorithm on several test cases.

Clustering by fast search-and-find of density peaks

Rodriguez Alex, A.;Laio, Alessandro
2014-01-01

Abstract

Cluster analysis is aimed at classifying elements into categories on the basis of their similarity. Its applications range from astronomy to bioinformatics, bibliometrics, and pattern recognition.We propose an approach based on the idea that cluster centers are characterized by a higher density than their neighbors and by a relatively large distance from points with higher densities. This idea forms the basis of a clustering procedure in which the number of clusters arises intuitively, outliers are automatically spotted and excluded fromthe analysis, and clusters are recognized regardless of their shape and of the dimensionality of the space inwhich they are embedded.We demonstrate the power of the algorithm on several test cases.
2014
344
6191
1492
1496
Rodriguez Alex, A.; Laio, Alessandro
File in questo prodotto:
File Dimensione Formato  
Science-2014-Rodriguez-1492-6.pdf

non disponibili

Tipologia: Versione Editoriale (PDF)
Licenza: Non specificato
Dimensione 1.89 MB
Formato Adobe PDF
1.89 MB Adobe PDF   Visualizza/Apri   Richiedi una copia
Laio-talk-1.pdf

accesso aperto

Tipologia: Altro materiale allegato
Licenza: Non specificato
Dimensione 9.92 MB
Formato Adobe PDF
9.92 MB Adobe PDF Visualizza/Apri

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/20.500.11767/14183
Citazioni
  • ???jsp.display-item.citation.pmc??? 351
  • Scopus 4265
  • ???jsp.display-item.citation.isi??? 3482
social impact