Spectral clustering of large-scale communities via random sketching and validation

Panagiotis A. Traganitis; Konstantinos Slavakis; Georgios B Giannakis

doi:10.1109/CISS.2015.7086867

Spectral clustering of large-scale communities via random sketching and validation

Panagiotis A. Traganitis, Konstantinos Slavakis, Georgios B Giannakis

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

2 Scopus citations

Abstract

In our era of data deluge, clustering algorithms that do not scale well with the dramatically increasing number of data have to be reconsidered. Spectral clustering, while powerful, is computationally and memory demanding, even for high performance computers. Capitalizing on the relationship between spectral clustering and kernel k-means, the present paper introduces a randomized algorithm for identifying communities in large-scale graphs based on a random sketching and validation approach, that enjoys reduced complexity compared to the clairvoyant spectral clustering. Numerical tests on synthetic and real data demonstrate the potential of the proposed approach.

Original language	English (US)
Title of host publication	2015 49th Annual Conference on Information Sciences and Systems, CISS 2015
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9781479984282
DOIs	https://doi.org/10.1109/CISS.2015.7086867
State	Published - Apr 15 2015
Event	2015 49th Annual Conference on Information Sciences and Systems, CISS 2015 - Baltimore, United States Duration: Mar 18 2015 → Mar 20 2015

Publication series

Name	2015 49th Annual Conference on Information Sciences and Systems, CISS 2015

Other

Other	2015 49th Annual Conference on Information Sciences and Systems, CISS 2015
Country/Territory	United States
City	Baltimore
Period	3/18/15 → 3/20/15

Bibliographical note

Publisher Copyright:
© 2015 IEEE.

Keywords

Sketch and validate
Spectral clustering
community identification

Access

10.1109/CISS.2015.7086867

OpenUrl availability

Full text

Cite this

Traganitis, P. A., Slavakis, K., & Giannakis, G. B. (2015). Spectral clustering of large-scale communities via random sketching and validation. In 2015 49th Annual Conference on Information Sciences and Systems, CISS 2015 Article 7086867 (2015 49th Annual Conference on Information Sciences and Systems, CISS 2015). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/CISS.2015.7086867

Spectral clustering of large-scale communities via random sketching and validation. / Traganitis, Panagiotis A.; Slavakis, Konstantinos; Giannakis, Georgios B.
2015 49th Annual Conference on Information Sciences and Systems, CISS 2015. Institute of Electrical and Electronics Engineers Inc., 2015. 7086867 (2015 49th Annual Conference on Information Sciences and Systems, CISS 2015).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Traganitis, PA, Slavakis, K & Giannakis, GB 2015, Spectral clustering of large-scale communities via random sketching and validation. in 2015 49th Annual Conference on Information Sciences and Systems, CISS 2015., 7086867, 2015 49th Annual Conference on Information Sciences and Systems, CISS 2015, Institute of Electrical and Electronics Engineers Inc., 2015 49th Annual Conference on Information Sciences and Systems, CISS 2015, Baltimore, United States, 3/18/15. https://doi.org/10.1109/CISS.2015.7086867

Traganitis PA, Slavakis K, Giannakis GB. Spectral clustering of large-scale communities via random sketching and validation. In 2015 49th Annual Conference on Information Sciences and Systems, CISS 2015. Institute of Electrical and Electronics Engineers Inc. 2015. 7086867. (2015 49th Annual Conference on Information Sciences and Systems, CISS 2015). doi: 10.1109/CISS.2015.7086867

Traganitis, Panagiotis A. ; Slavakis, Konstantinos ; Giannakis, Georgios B. / Spectral clustering of large-scale communities via random sketching and validation. 2015 49th Annual Conference on Information Sciences and Systems, CISS 2015. Institute of Electrical and Electronics Engineers Inc., 2015. (2015 49th Annual Conference on Information Sciences and Systems, CISS 2015).

@inproceedings{ae1ed73431d14a43bd21d9768d42cd6a,

title = "Spectral clustering of large-scale communities via random sketching and validation",

abstract = "In our era of data deluge, clustering algorithms that do not scale well with the dramatically increasing number of data have to be reconsidered. Spectral clustering, while powerful, is computationally and memory demanding, even for high performance computers. Capitalizing on the relationship between spectral clustering and kernel k-means, the present paper introduces a randomized algorithm for identifying communities in large-scale graphs based on a random sketching and validation approach, that enjoys reduced complexity compared to the clairvoyant spectral clustering. Numerical tests on synthetic and real data demonstrate the potential of the proposed approach.",

keywords = "Sketch and validate, Spectral clustering, community identification",

author = "Traganitis, {Panagiotis A.} and Konstantinos Slavakis and Giannakis, {Georgios B}",

note = "Publisher Copyright: {\textcopyright} 2015 IEEE.; 2015 49th Annual Conference on Information Sciences and Systems, CISS 2015 ; Conference date: 18-03-2015 Through 20-03-2015",

year = "2015",

month = apr,

day = "15",

doi = "10.1109/CISS.2015.7086867",

language = "English (US)",

series = "2015 49th Annual Conference on Information Sciences and Systems, CISS 2015",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

booktitle = "2015 49th Annual Conference on Information Sciences and Systems, CISS 2015",

}

TY - GEN

T1 - Spectral clustering of large-scale communities via random sketching and validation

AU - Traganitis, Panagiotis A.

AU - Slavakis, Konstantinos

AU - Giannakis, Georgios B

PY - 2015/4/15

Y1 - 2015/4/15

N2 - In our era of data deluge, clustering algorithms that do not scale well with the dramatically increasing number of data have to be reconsidered. Spectral clustering, while powerful, is computationally and memory demanding, even for high performance computers. Capitalizing on the relationship between spectral clustering and kernel k-means, the present paper introduces a randomized algorithm for identifying communities in large-scale graphs based on a random sketching and validation approach, that enjoys reduced complexity compared to the clairvoyant spectral clustering. Numerical tests on synthetic and real data demonstrate the potential of the proposed approach.

AB - In our era of data deluge, clustering algorithms that do not scale well with the dramatically increasing number of data have to be reconsidered. Spectral clustering, while powerful, is computationally and memory demanding, even for high performance computers. Capitalizing on the relationship between spectral clustering and kernel k-means, the present paper introduces a randomized algorithm for identifying communities in large-scale graphs based on a random sketching and validation approach, that enjoys reduced complexity compared to the clairvoyant spectral clustering. Numerical tests on synthetic and real data demonstrate the potential of the proposed approach.

KW - Sketch and validate

KW - Spectral clustering

KW - community identification

UR - http://www.scopus.com/inward/record.url?scp=84929190841&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84929190841&partnerID=8YFLogxK

U2 - 10.1109/CISS.2015.7086867

DO - 10.1109/CISS.2015.7086867

M3 - Conference contribution

AN - SCOPUS:84929190841

T3 - 2015 49th Annual Conference on Information Sciences and Systems, CISS 2015

BT - 2015 49th Annual Conference on Information Sciences and Systems, CISS 2015

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2015 49th Annual Conference on Information Sciences and Systems, CISS 2015

Y2 - 18 March 2015 through 20 March 2015

ER -

Spectral clustering of large-scale communities via random sketching and validation

Abstract

Publication series

Other

Bibliographical note

Keywords

Access

OpenUrl availability

Other files and links

Fingerprint

Cite this