Agglomerative hierarchical kernel spectral data clustering

RaghvenPhDa Mall, Rocco Langone, Johan A.K. Suykens

Research output: Chapter in Book/Report/Conference proceedingConference contribution

4 Citations (Scopus)

Abstract

In this paper we extend the agglomerative hierarchical kernel spectral clustering (AH-KSC [1]) technique from networks to datasets and images. The kernel spectral clustering (KSC) technique builds a clustering model in a primal-dual optimization framework. The dual solution leads to an eigen-decomposition. The clustering model consists of kernel evaluations, projections onto the eigenvectors and a powerful out-of-sample extension property. We first estimate the optimal model parameters using the balanced angular fitting (BAF) [2] criterion. We then exploit the eigen-projections corresponding to these parameters to automatically identify a set of increasing distance thresholds. These distance thresholds provide the clusters at different levels of hierarchy in the dataset which are merged in an agglomerative fashion as shown in [1], [4]. We showcase the effectiveness of the AH-KSC method on several datasets and real world images. We compare the AH-KSC method with several agglomerative hierarchical clustering techniques and overcome the issues of hierarchical KSC technique proposed in [5].

Original languageEnglish
Title of host publicationIEEE SSCI 2014 - 2014 IEEE Symposium Series on Computational Intelligence - CIDM 2014
Subtitle of host publication2014 IEEE Symposium on Computational Intelligence and Data Mining, Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages9-16
Number of pages8
ISBN (Electronic)9781479945191
DOIs
Publication statusPublished - 1 Jan 2015
Externally publishedYes
Event5th IEEE Symposium on Computational Intelligence and Data Mining, CIDM 2014 - Orlando, United States
Duration: 9 Dec 201412 Dec 2014

Other

Other5th IEEE Symposium on Computational Intelligence and Data Mining, CIDM 2014
CountryUnited States
CityOrlando
Period9/12/1412/12/14

Fingerprint

Eigenvalues and eigenfunctions
Decomposition

ASJC Scopus subject areas

  • Artificial Intelligence
  • Information Systems
  • Signal Processing
  • Software

Cite this

Mall, R., Langone, R., & Suykens, J. A. K. (2015). Agglomerative hierarchical kernel spectral data clustering. In IEEE SSCI 2014 - 2014 IEEE Symposium Series on Computational Intelligence - CIDM 2014: 2014 IEEE Symposium on Computational Intelligence and Data Mining, Proceedings (pp. 9-16). [7008142] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/CIDM.2014.7008142

Agglomerative hierarchical kernel spectral data clustering. / Mall, RaghvenPhDa; Langone, Rocco; Suykens, Johan A.K.

IEEE SSCI 2014 - 2014 IEEE Symposium Series on Computational Intelligence - CIDM 2014: 2014 IEEE Symposium on Computational Intelligence and Data Mining, Proceedings. Institute of Electrical and Electronics Engineers Inc., 2015. p. 9-16 7008142.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Mall, R, Langone, R & Suykens, JAK 2015, Agglomerative hierarchical kernel spectral data clustering. in IEEE SSCI 2014 - 2014 IEEE Symposium Series on Computational Intelligence - CIDM 2014: 2014 IEEE Symposium on Computational Intelligence and Data Mining, Proceedings., 7008142, Institute of Electrical and Electronics Engineers Inc., pp. 9-16, 5th IEEE Symposium on Computational Intelligence and Data Mining, CIDM 2014, Orlando, United States, 9/12/14. https://doi.org/10.1109/CIDM.2014.7008142
Mall R, Langone R, Suykens JAK. Agglomerative hierarchical kernel spectral data clustering. In IEEE SSCI 2014 - 2014 IEEE Symposium Series on Computational Intelligence - CIDM 2014: 2014 IEEE Symposium on Computational Intelligence and Data Mining, Proceedings. Institute of Electrical and Electronics Engineers Inc. 2015. p. 9-16. 7008142 https://doi.org/10.1109/CIDM.2014.7008142
Mall, RaghvenPhDa ; Langone, Rocco ; Suykens, Johan A.K. / Agglomerative hierarchical kernel spectral data clustering. IEEE SSCI 2014 - 2014 IEEE Symposium Series on Computational Intelligence - CIDM 2014: 2014 IEEE Symposium on Computational Intelligence and Data Mining, Proceedings. Institute of Electrical and Electronics Engineers Inc., 2015. pp. 9-16
@inproceedings{13ca7071debe4fedbd0b900d7e4e726d,
title = "Agglomerative hierarchical kernel spectral data clustering",
abstract = "In this paper we extend the agglomerative hierarchical kernel spectral clustering (AH-KSC [1]) technique from networks to datasets and images. The kernel spectral clustering (KSC) technique builds a clustering model in a primal-dual optimization framework. The dual solution leads to an eigen-decomposition. The clustering model consists of kernel evaluations, projections onto the eigenvectors and a powerful out-of-sample extension property. We first estimate the optimal model parameters using the balanced angular fitting (BAF) [2] criterion. We then exploit the eigen-projections corresponding to these parameters to automatically identify a set of increasing distance thresholds. These distance thresholds provide the clusters at different levels of hierarchy in the dataset which are merged in an agglomerative fashion as shown in [1], [4]. We showcase the effectiveness of the AH-KSC method on several datasets and real world images. We compare the AH-KSC method with several agglomerative hierarchical clustering techniques and overcome the issues of hierarchical KSC technique proposed in [5].",
author = "RaghvenPhDa Mall and Rocco Langone and Suykens, {Johan A.K.}",
year = "2015",
month = "1",
day = "1",
doi = "10.1109/CIDM.2014.7008142",
language = "English",
pages = "9--16",
booktitle = "IEEE SSCI 2014 - 2014 IEEE Symposium Series on Computational Intelligence - CIDM 2014",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Agglomerative hierarchical kernel spectral data clustering

AU - Mall, RaghvenPhDa

AU - Langone, Rocco

AU - Suykens, Johan A.K.

PY - 2015/1/1

Y1 - 2015/1/1

N2 - In this paper we extend the agglomerative hierarchical kernel spectral clustering (AH-KSC [1]) technique from networks to datasets and images. The kernel spectral clustering (KSC) technique builds a clustering model in a primal-dual optimization framework. The dual solution leads to an eigen-decomposition. The clustering model consists of kernel evaluations, projections onto the eigenvectors and a powerful out-of-sample extension property. We first estimate the optimal model parameters using the balanced angular fitting (BAF) [2] criterion. We then exploit the eigen-projections corresponding to these parameters to automatically identify a set of increasing distance thresholds. These distance thresholds provide the clusters at different levels of hierarchy in the dataset which are merged in an agglomerative fashion as shown in [1], [4]. We showcase the effectiveness of the AH-KSC method on several datasets and real world images. We compare the AH-KSC method with several agglomerative hierarchical clustering techniques and overcome the issues of hierarchical KSC technique proposed in [5].

AB - In this paper we extend the agglomerative hierarchical kernel spectral clustering (AH-KSC [1]) technique from networks to datasets and images. The kernel spectral clustering (KSC) technique builds a clustering model in a primal-dual optimization framework. The dual solution leads to an eigen-decomposition. The clustering model consists of kernel evaluations, projections onto the eigenvectors and a powerful out-of-sample extension property. We first estimate the optimal model parameters using the balanced angular fitting (BAF) [2] criterion. We then exploit the eigen-projections corresponding to these parameters to automatically identify a set of increasing distance thresholds. These distance thresholds provide the clusters at different levels of hierarchy in the dataset which are merged in an agglomerative fashion as shown in [1], [4]. We showcase the effectiveness of the AH-KSC method on several datasets and real world images. We compare the AH-KSC method with several agglomerative hierarchical clustering techniques and overcome the issues of hierarchical KSC technique proposed in [5].

UR - http://www.scopus.com/inward/record.url?scp=84925071196&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84925071196&partnerID=8YFLogxK

U2 - 10.1109/CIDM.2014.7008142

DO - 10.1109/CIDM.2014.7008142

M3 - Conference contribution

AN - SCOPUS:84925071196

SP - 9

EP - 16

BT - IEEE SSCI 2014 - 2014 IEEE Symposium Series on Computational Intelligence - CIDM 2014

PB - Institute of Electrical and Electronics Engineers Inc.

ER -