Accuracy and robustness of clustering algorithms for small-size applications in bioinformatics

Pamela Minicozzi; Fabio Rapallo; Enrico Scalas; Francesco Dondero

doi:10.1016/j.physa.2008.07.026

Accuracy and robustness of clustering algorithms for small-size applications in bioinformatics

Pamela Minicozzi
, Fabio Rapallo
, Enrico Scalas
, Francesco Dondero

Dipartimento di Scienze e Innovazione Tecnologica

Risultato della ricerca: Contributo su rivista › Articolo in rivista › peer review

Abstract

The performance (accuracy and robustness) of several clustering algorithms is studied for linearly dependent random variables in the presence of noise. It turns out that the error percentage quickly increases when the number of observations is less than the number of variables. This situation is common situation in experiments with DNA microarrays. Moreover, an a posteriori criterion to choose between two discordant clustering algorithm is presented.

Lingua originale	Inglese
pagine (da-a)	6310-6318
Numero di pagine	9
Rivista	Physica A: Statistical Mechanics and its Applications
Volume	387
Numero di pubblicazione	25
DOI	https://doi.org/10.1016/j.physa.2008.07.026
Stato di pubblicazione	Pubblicato - 1 nov 2008

Accesso al documento

10.1016/j.physa.2008.07.026

Altri file e collegamenti

Link to publication in Scopus

Cita questo

@article{c72553d6d5334ec4ba9ce6975c004753,

title = "Accuracy and robustness of clustering algorithms for small-size applications in bioinformatics",

abstract = "The performance (accuracy and robustness) of several clustering algorithms is studied for linearly dependent random variables in the presence of noise. It turns out that the error percentage quickly increases when the number of observations is less than the number of variables. This situation is common situation in experiments with DNA microarrays. Moreover, an a posteriori criterion to choose between two discordant clustering algorithm is presented.",

keywords = "Accuracy, Clustering, DNA microarray, Robustness",

author = "Pamela Minicozzi and Fabio Rapallo and Enrico Scalas and Francesco Dondero",

note = "Funding Information: This paper is part of a research project devoted to understanding the effect of noise in unsupervised clustering methods and was also motivated by measurements carried out using microarrays on different environmental relevant model species exposed to adverse conditions, see Refs. [18–20] . This work has been supported by two Italian grants: MIUR PRIN 2006 (Minicozzi, Scalas, Dondero) and UPO “Ricerca Locale” 2007 (Rapallo).",

year = "2008",

month = nov,

day = "1",

doi = "10.1016/j.physa.2008.07.026",

language = "English",

volume = "387",

pages = "6310--6318",

journal = "Physica A: Statistical Mechanics and its Applications",

issn = "0378-4371",

publisher = "Elsevier BV",

number = "25",

}

TY - JOUR

T1 - Accuracy and robustness of clustering algorithms for small-size applications in bioinformatics

AU - Minicozzi, Pamela

AU - Rapallo, Fabio

AU - Scalas, Enrico

AU - Dondero, Francesco

N1 - Funding Information: This paper is part of a research project devoted to understanding the effect of noise in unsupervised clustering methods and was also motivated by measurements carried out using microarrays on different environmental relevant model species exposed to adverse conditions, see Refs. [18–20] . This work has been supported by two Italian grants: MIUR PRIN 2006 (Minicozzi, Scalas, Dondero) and UPO “Ricerca Locale” 2007 (Rapallo).

PY - 2008/11/1

Y1 - 2008/11/1

N2 - The performance (accuracy and robustness) of several clustering algorithms is studied for linearly dependent random variables in the presence of noise. It turns out that the error percentage quickly increases when the number of observations is less than the number of variables. This situation is common situation in experiments with DNA microarrays. Moreover, an a posteriori criterion to choose between two discordant clustering algorithm is presented.

AB - The performance (accuracy and robustness) of several clustering algorithms is studied for linearly dependent random variables in the presence of noise. It turns out that the error percentage quickly increases when the number of observations is less than the number of variables. This situation is common situation in experiments with DNA microarrays. Moreover, an a posteriori criterion to choose between two discordant clustering algorithm is presented.

KW - Accuracy

KW - Clustering

KW - DNA microarray

KW - Robustness

UR - https://www.scopus.com/pages/publications/50949119718

U2 - 10.1016/j.physa.2008.07.026

DO - 10.1016/j.physa.2008.07.026

M3 - Article

SN - 0378-4371

VL - 387

SP - 6310

EP - 6318

JO - Physica A: Statistical Mechanics and its Applications

JF - Physica A: Statistical Mechanics and its Applications

IS - 25

ER -

Accuracy and robustness of clustering algorithms for small-size applications in bioinformatics

Abstract

Accesso al documento

Altri file e collegamenti

Fingerprint

Cita questo