Consensus clustering using partial evidence accumulation

A. Lourenço, S.R. Bulò, Nicola Rebagliati, A. Fred, M. Figueiredo, M. Pelillo

Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingsScientificpeer-review

6 Citations (Scopus)

Abstract

The Evidence Accumulation Clustering, EAC, algorithm is a clustering ensemble method which uses co-occurrence statistics to derive a similarity matrix, referred to as co-association matrix. In order to obtain a final consensus clustering the co-association matrix is fed to a pairwise similarity clustering algorithm. The method has thus O(n 2) space complexity, which can constitute a relevant bottleneck to its scalability. In this paper we propose a new formulation which works using a partial set of the co-occurrences, greatly reducing the computational time and space, leading to a scalable algorithm. Experiments on both synthetic and real benchmark data show the effectiveness of the proposed approach.
Original languageEnglish
Title of host publicationPattern Recognition and Image Analysis
Subtitle of host publicationIbPRIA 2013
EditorsJ.M. Sanches, L. Mico, J.S. Cardoso
Place of PublicationBerlin, Heidelberg
PublisherSpringer
Pages69-78
ISBN (Electronic)978-3-642-38628-2
ISBN (Print)978-3-642-38627-5
DOIs
Publication statusPublished - 2013
MoE publication typeNot Eligible
Event6th Iberian Conference on Pattern Recognition and Image Analysis, IbPRIA 2013 - Funchal, Madeira, Portugal
Duration: 5 Jun 20137 Jun 2013

Publication series

SeriesLecture Notes in Computer Science
Volume7887
ISSN0302-9743

Conference

Conference6th Iberian Conference on Pattern Recognition and Image Analysis, IbPRIA 2013
Abbreviated titleIbPRIA 2013
CountryPortugal
CityFunchal, Madeira
Period5/06/137/06/13

Fingerprint

Clustering algorithms
Scalability
Statistics
Experiments

Keywords

  • clustering algorithm
  • clustering ensembles
  • evidence accumulation clustering
  • scalable algorithms

Cite this

Lourenço, A., Bulò, S. R., Rebagliati, N., Fred, A., Figueiredo, M., & Pelillo, M. (2013). Consensus clustering using partial evidence accumulation. In J. M. Sanches, L. Mico, & J. S. Cardoso (Eds.), Pattern Recognition and Image Analysis: IbPRIA 2013 (pp. 69-78). Berlin, Heidelberg: Springer. Lecture Notes in Computer Science, Vol.. 7887 https://doi.org/10.1007/978-3-642-38628-2_8
Lourenço, A. ; Bulò, S.R. ; Rebagliati, Nicola ; Fred, A. ; Figueiredo, M. ; Pelillo, M. / Consensus clustering using partial evidence accumulation. Pattern Recognition and Image Analysis: IbPRIA 2013. editor / J.M. Sanches ; L. Mico ; J.S. Cardoso. Berlin, Heidelberg : Springer, 2013. pp. 69-78 (Lecture Notes in Computer Science, Vol. 7887 ).
@inproceedings{39b94a7925a24f5c8a5aa17293fb3159,
title = "Consensus clustering using partial evidence accumulation",
abstract = "The Evidence Accumulation Clustering, EAC, algorithm is a clustering ensemble method which uses co-occurrence statistics to derive a similarity matrix, referred to as co-association matrix. In order to obtain a final consensus clustering the co-association matrix is fed to a pairwise similarity clustering algorithm. The method has thus O(n 2) space complexity, which can constitute a relevant bottleneck to its scalability. In this paper we propose a new formulation which works using a partial set of the co-occurrences, greatly reducing the computational time and space, leading to a scalable algorithm. Experiments on both synthetic and real benchmark data show the effectiveness of the proposed approach.",
keywords = "clustering algorithm, clustering ensembles, evidence accumulation clustering, scalable algorithms",
author = "A. Louren{\cc}o and S.R. Bul{\`o} and Nicola Rebagliati and A. Fred and M. Figueiredo and M. Pelillo",
year = "2013",
doi = "10.1007/978-3-642-38628-2_8",
language = "English",
isbn = "978-3-642-38627-5",
series = "Lecture Notes in Computer Science",
publisher = "Springer",
pages = "69--78",
editor = "J.M. Sanches and L. Mico and J.S. Cardoso",
booktitle = "Pattern Recognition and Image Analysis",
address = "Germany",

}

Lourenço, A, Bulò, SR, Rebagliati, N, Fred, A, Figueiredo, M & Pelillo, M 2013, Consensus clustering using partial evidence accumulation. in JM Sanches, L Mico & JS Cardoso (eds), Pattern Recognition and Image Analysis: IbPRIA 2013. Springer, Berlin, Heidelberg, Lecture Notes in Computer Science, vol. 7887 , pp. 69-78, 6th Iberian Conference on Pattern Recognition and Image Analysis, IbPRIA 2013, Funchal, Madeira, Portugal, 5/06/13. https://doi.org/10.1007/978-3-642-38628-2_8

Consensus clustering using partial evidence accumulation. / Lourenço, A.; Bulò, S.R.; Rebagliati, Nicola; Fred, A.; Figueiredo, M.; Pelillo, M.

Pattern Recognition and Image Analysis: IbPRIA 2013. ed. / J.M. Sanches; L. Mico; J.S. Cardoso. Berlin, Heidelberg : Springer, 2013. p. 69-78 (Lecture Notes in Computer Science, Vol. 7887 ).

Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingsScientificpeer-review

TY - GEN

T1 - Consensus clustering using partial evidence accumulation

AU - Lourenço, A.

AU - Bulò, S.R.

AU - Rebagliati, Nicola

AU - Fred, A.

AU - Figueiredo, M.

AU - Pelillo, M.

PY - 2013

Y1 - 2013

N2 - The Evidence Accumulation Clustering, EAC, algorithm is a clustering ensemble method which uses co-occurrence statistics to derive a similarity matrix, referred to as co-association matrix. In order to obtain a final consensus clustering the co-association matrix is fed to a pairwise similarity clustering algorithm. The method has thus O(n 2) space complexity, which can constitute a relevant bottleneck to its scalability. In this paper we propose a new formulation which works using a partial set of the co-occurrences, greatly reducing the computational time and space, leading to a scalable algorithm. Experiments on both synthetic and real benchmark data show the effectiveness of the proposed approach.

AB - The Evidence Accumulation Clustering, EAC, algorithm is a clustering ensemble method which uses co-occurrence statistics to derive a similarity matrix, referred to as co-association matrix. In order to obtain a final consensus clustering the co-association matrix is fed to a pairwise similarity clustering algorithm. The method has thus O(n 2) space complexity, which can constitute a relevant bottleneck to its scalability. In this paper we propose a new formulation which works using a partial set of the co-occurrences, greatly reducing the computational time and space, leading to a scalable algorithm. Experiments on both synthetic and real benchmark data show the effectiveness of the proposed approach.

KW - clustering algorithm

KW - clustering ensembles

KW - evidence accumulation clustering

KW - scalable algorithms

U2 - 10.1007/978-3-642-38628-2_8

DO - 10.1007/978-3-642-38628-2_8

M3 - Conference article in proceedings

SN - 978-3-642-38627-5

T3 - Lecture Notes in Computer Science

SP - 69

EP - 78

BT - Pattern Recognition and Image Analysis

A2 - Sanches, J.M.

A2 - Mico, L.

A2 - Cardoso, J.S.

PB - Springer

CY - Berlin, Heidelberg

ER -

Lourenço A, Bulò SR, Rebagliati N, Fred A, Figueiredo M, Pelillo M. Consensus clustering using partial evidence accumulation. In Sanches JM, Mico L, Cardoso JS, editors, Pattern Recognition and Image Analysis: IbPRIA 2013. Berlin, Heidelberg: Springer. 2013. p. 69-78. (Lecture Notes in Computer Science, Vol. 7887 ). https://doi.org/10.1007/978-3-642-38628-2_8