Multiclass Semisupervised Learning Based Upon Kernel Spectral Clustering

Siamak Mehrkanoon, Carlos Alzate, RaghvenPhDa Mall, Rocco Langone, Johan A.K. Suykens

Research output: Contribution to journalArticle

30 Citations (Scopus)

Abstract

This paper proposes a multiclass semisupervised learning algorithm by using kernel spectral clustering (KSC) as a core model. A regularized KSC is formulated to estimate the class memberships of data points in a semisupervised setting using the one-versus-all strategy while both labeled and unlabeled data points are present in the learning process. The propagation of the labels to a large amount of unlabeled data points is achieved by adding the regularization terms to the cost function of the KSC formulation. In other words, imposing the regularization term enforces certain desired memberships. The model is then obtained by solving a linear system in the dual. Furthermore, the optimal embedding dimension is designed for semisupervised clustering. This plays a key role when one deals with a large number of clusters.

Original languageEnglish
Article number6822553
Pages (from-to)720-733
Number of pages14
JournalIEEE Transactions on Neural Networks and Learning Systems
Volume26
Issue number4
DOIs
Publication statusPublished - 1 Apr 2015
Externally publishedYes

Fingerprint

Cost functions
Learning algorithms
Linear systems
Labels

Keywords

  • Kernel spectral clustering (KSC)
  • low embedding dimension for clustering
  • multiclass problem
  • semisupervised learning

ASJC Scopus subject areas

  • Software
  • Computer Science Applications
  • Computer Networks and Communications
  • Artificial Intelligence

Cite this

Multiclass Semisupervised Learning Based Upon Kernel Spectral Clustering. / Mehrkanoon, Siamak; Alzate, Carlos; Mall, RaghvenPhDa; Langone, Rocco; Suykens, Johan A.K.

In: IEEE Transactions on Neural Networks and Learning Systems, Vol. 26, No. 4, 6822553, 01.04.2015, p. 720-733.

Research output: Contribution to journalArticle

Mehrkanoon, Siamak ; Alzate, Carlos ; Mall, RaghvenPhDa ; Langone, Rocco ; Suykens, Johan A.K. / Multiclass Semisupervised Learning Based Upon Kernel Spectral Clustering. In: IEEE Transactions on Neural Networks and Learning Systems. 2015 ; Vol. 26, No. 4. pp. 720-733.
@article{9ff864e0808a43078372a6cb2aa3b48e,
title = "Multiclass Semisupervised Learning Based Upon Kernel Spectral Clustering",
abstract = "This paper proposes a multiclass semisupervised learning algorithm by using kernel spectral clustering (KSC) as a core model. A regularized KSC is formulated to estimate the class memberships of data points in a semisupervised setting using the one-versus-all strategy while both labeled and unlabeled data points are present in the learning process. The propagation of the labels to a large amount of unlabeled data points is achieved by adding the regularization terms to the cost function of the KSC formulation. In other words, imposing the regularization term enforces certain desired memberships. The model is then obtained by solving a linear system in the dual. Furthermore, the optimal embedding dimension is designed for semisupervised clustering. This plays a key role when one deals with a large number of clusters.",
keywords = "Kernel spectral clustering (KSC), low embedding dimension for clustering, multiclass problem, semisupervised learning",
author = "Siamak Mehrkanoon and Carlos Alzate and RaghvenPhDa Mall and Rocco Langone and Suykens, {Johan A.K.}",
year = "2015",
month = "4",
day = "1",
doi = "10.1109/TNNLS.2014.2322377",
language = "English",
volume = "26",
pages = "720--733",
journal = "IEEE Transactions on Neural Networks and Learning Systems",
issn = "2162-237X",
publisher = "IEEE Computational Intelligence Society",
number = "4",

}

TY - JOUR

T1 - Multiclass Semisupervised Learning Based Upon Kernel Spectral Clustering

AU - Mehrkanoon, Siamak

AU - Alzate, Carlos

AU - Mall, RaghvenPhDa

AU - Langone, Rocco

AU - Suykens, Johan A.K.

PY - 2015/4/1

Y1 - 2015/4/1

N2 - This paper proposes a multiclass semisupervised learning algorithm by using kernel spectral clustering (KSC) as a core model. A regularized KSC is formulated to estimate the class memberships of data points in a semisupervised setting using the one-versus-all strategy while both labeled and unlabeled data points are present in the learning process. The propagation of the labels to a large amount of unlabeled data points is achieved by adding the regularization terms to the cost function of the KSC formulation. In other words, imposing the regularization term enforces certain desired memberships. The model is then obtained by solving a linear system in the dual. Furthermore, the optimal embedding dimension is designed for semisupervised clustering. This plays a key role when one deals with a large number of clusters.

AB - This paper proposes a multiclass semisupervised learning algorithm by using kernel spectral clustering (KSC) as a core model. A regularized KSC is formulated to estimate the class memberships of data points in a semisupervised setting using the one-versus-all strategy while both labeled and unlabeled data points are present in the learning process. The propagation of the labels to a large amount of unlabeled data points is achieved by adding the regularization terms to the cost function of the KSC formulation. In other words, imposing the regularization term enforces certain desired memberships. The model is then obtained by solving a linear system in the dual. Furthermore, the optimal embedding dimension is designed for semisupervised clustering. This plays a key role when one deals with a large number of clusters.

KW - Kernel spectral clustering (KSC)

KW - low embedding dimension for clustering

KW - multiclass problem

KW - semisupervised learning

UR - http://www.scopus.com/inward/record.url?scp=85027937419&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85027937419&partnerID=8YFLogxK

U2 - 10.1109/TNNLS.2014.2322377

DO - 10.1109/TNNLS.2014.2322377

M3 - Article

AN - SCOPUS:85027937419

VL - 26

SP - 720

EP - 733

JO - IEEE Transactions on Neural Networks and Learning Systems

JF - IEEE Transactions on Neural Networks and Learning Systems

SN - 2162-237X

IS - 4

M1 - 6822553

ER -