Hierarchical semi-supervised clustering using KSC based model

Siamak Mehrkanoon, Oscar Mauricio Agudelo, RaghvenPhDa Mall, Johan A.K. Suykens

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

Abstract

This paper introduces a methodology to incorporate the label information in discovering the underlying clusters in a hierarchical setting using multi-class semi-supervised clustering algorithm. The method aims at revealing the relationship between clusters given few labels associated to some of the clusters. The problem is formulated as a regularized kernel spectral clustering algorithm in the primal-dual setting. The available labels are incorporated in different levels of hierarchy from top to bottom. As we advance towards the lowers levels in the tree all the previously added labels are used in the generation of the new levels of hierarchy. The model is trained on a subset of the data and then applied to the rest of the data in a learning framework. Thanks to the previously learned model, the out-of-sample extension property of the model allows then to predict the memberships of a new point. A combination of an internal clustering quality index and classification accuracy is used for model selection. Experiments are conducted on synthetic data and real image segmentation problems to show the applicability of the proposed approach.

Original languageEnglish
Title of host publication2015 International Joint Conference on Neural Networks, IJCNN 2015
PublisherInstitute of Electrical and Electronics Engineers Inc.
Volume2015-September
ISBN (Electronic)9781479919604, 9781479919604, 9781479919604, 9781479919604
DOIs
Publication statusPublished - 28 Sep 2015
Externally publishedYes
EventInternational Joint Conference on Neural Networks, IJCNN 2015 - Killarney, Ireland
Duration: 12 Jul 201517 Jul 2015

Other

OtherInternational Joint Conference on Neural Networks, IJCNN 2015
CountryIreland
CityKillarney
Period12/7/1517/7/15

Fingerprint

Labels
Clustering algorithms
Image segmentation
Experiments

ASJC Scopus subject areas

  • Software
  • Artificial Intelligence

Cite this

Mehrkanoon, S., Agudelo, O. M., Mall, R., & Suykens, J. A. K. (2015). Hierarchical semi-supervised clustering using KSC based model. In 2015 International Joint Conference on Neural Networks, IJCNN 2015 (Vol. 2015-September). [7280682] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IJCNN.2015.7280682

Hierarchical semi-supervised clustering using KSC based model. / Mehrkanoon, Siamak; Agudelo, Oscar Mauricio; Mall, RaghvenPhDa; Suykens, Johan A.K.

2015 International Joint Conference on Neural Networks, IJCNN 2015. Vol. 2015-September Institute of Electrical and Electronics Engineers Inc., 2015. 7280682.

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

Mehrkanoon, S, Agudelo, OM, Mall, R & Suykens, JAK 2015, Hierarchical semi-supervised clustering using KSC based model. in 2015 International Joint Conference on Neural Networks, IJCNN 2015. vol. 2015-September, 7280682, Institute of Electrical and Electronics Engineers Inc., International Joint Conference on Neural Networks, IJCNN 2015, Killarney, Ireland, 12/7/15. https://doi.org/10.1109/IJCNN.2015.7280682
Mehrkanoon S, Agudelo OM, Mall R, Suykens JAK. Hierarchical semi-supervised clustering using KSC based model. In 2015 International Joint Conference on Neural Networks, IJCNN 2015. Vol. 2015-September. Institute of Electrical and Electronics Engineers Inc. 2015. 7280682 https://doi.org/10.1109/IJCNN.2015.7280682
Mehrkanoon, Siamak ; Agudelo, Oscar Mauricio ; Mall, RaghvenPhDa ; Suykens, Johan A.K. / Hierarchical semi-supervised clustering using KSC based model. 2015 International Joint Conference on Neural Networks, IJCNN 2015. Vol. 2015-September Institute of Electrical and Electronics Engineers Inc., 2015.
@inproceedings{ca5656e454ea4deb9db382e1afe35539,
title = "Hierarchical semi-supervised clustering using KSC based model",
abstract = "This paper introduces a methodology to incorporate the label information in discovering the underlying clusters in a hierarchical setting using multi-class semi-supervised clustering algorithm. The method aims at revealing the relationship between clusters given few labels associated to some of the clusters. The problem is formulated as a regularized kernel spectral clustering algorithm in the primal-dual setting. The available labels are incorporated in different levels of hierarchy from top to bottom. As we advance towards the lowers levels in the tree all the previously added labels are used in the generation of the new levels of hierarchy. The model is trained on a subset of the data and then applied to the rest of the data in a learning framework. Thanks to the previously learned model, the out-of-sample extension property of the model allows then to predict the memberships of a new point. A combination of an internal clustering quality index and classification accuracy is used for model selection. Experiments are conducted on synthetic data and real image segmentation problems to show the applicability of the proposed approach.",
author = "Siamak Mehrkanoon and Agudelo, {Oscar Mauricio} and RaghvenPhDa Mall and Suykens, {Johan A.K.}",
year = "2015",
month = "9",
day = "28",
doi = "10.1109/IJCNN.2015.7280682",
language = "English",
volume = "2015-September",
booktitle = "2015 International Joint Conference on Neural Networks, IJCNN 2015",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Hierarchical semi-supervised clustering using KSC based model

AU - Mehrkanoon, Siamak

AU - Agudelo, Oscar Mauricio

AU - Mall, RaghvenPhDa

AU - Suykens, Johan A.K.

PY - 2015/9/28

Y1 - 2015/9/28

N2 - This paper introduces a methodology to incorporate the label information in discovering the underlying clusters in a hierarchical setting using multi-class semi-supervised clustering algorithm. The method aims at revealing the relationship between clusters given few labels associated to some of the clusters. The problem is formulated as a regularized kernel spectral clustering algorithm in the primal-dual setting. The available labels are incorporated in different levels of hierarchy from top to bottom. As we advance towards the lowers levels in the tree all the previously added labels are used in the generation of the new levels of hierarchy. The model is trained on a subset of the data and then applied to the rest of the data in a learning framework. Thanks to the previously learned model, the out-of-sample extension property of the model allows then to predict the memberships of a new point. A combination of an internal clustering quality index and classification accuracy is used for model selection. Experiments are conducted on synthetic data and real image segmentation problems to show the applicability of the proposed approach.

AB - This paper introduces a methodology to incorporate the label information in discovering the underlying clusters in a hierarchical setting using multi-class semi-supervised clustering algorithm. The method aims at revealing the relationship between clusters given few labels associated to some of the clusters. The problem is formulated as a regularized kernel spectral clustering algorithm in the primal-dual setting. The available labels are incorporated in different levels of hierarchy from top to bottom. As we advance towards the lowers levels in the tree all the previously added labels are used in the generation of the new levels of hierarchy. The model is trained on a subset of the data and then applied to the rest of the data in a learning framework. Thanks to the previously learned model, the out-of-sample extension property of the model allows then to predict the memberships of a new point. A combination of an internal clustering quality index and classification accuracy is used for model selection. Experiments are conducted on synthetic data and real image segmentation problems to show the applicability of the proposed approach.

UR - http://www.scopus.com/inward/record.url?scp=84951000752&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84951000752&partnerID=8YFLogxK

U2 - 10.1109/IJCNN.2015.7280682

DO - 10.1109/IJCNN.2015.7280682

M3 - Conference contribution

VL - 2015-September

BT - 2015 International Joint Conference on Neural Networks, IJCNN 2015

PB - Institute of Electrical and Electronics Engineers Inc.

ER -