Optimized data fusion for K-means Laplacian clustering

Shi Yu, Xinhai Liu, Léon Charles Tranchevent, W. Glänzel, Johan A K Suykens, Bart de Moor, Yves Moreau

Research output: Contribution to journalArticle

13 Citations (Scopus)

Abstract

Motivation: We propose a novel algorithm to combine multiple kernels and Laplacians for clustering analysis. The new algorithm is formulated on a Rayleigh quotient objective function and is solved as a bi-level alternating minimization procedure. Using the proposed algorithm, the coefficients of kernels and Laplacians can be optimized automatically. Results: Three variants of the algorithm are proposed. The performance is systematically validated on two real-life data fusion applications. The proposed Optimized Kernel Laplacian Clustering (OKLC) algorithms perform significantly better than other methods. Moreover, the coefficients of kernels and Laplacians optimized by OKLC show some correlation with the rank of performance of individual data source. Though in our evaluation the K values are predefined, in practical studies, the optimal cluster number can be consistently estimated from the eigenspectrum of the combined kernel Laplacian matrix.

Original languageEnglish
Article numberbtq569
Pages (from-to)118-126
Number of pages9
JournalBioinformatics
Volume27
Issue number1
DOIs
Publication statusPublished - Jan 2011

Fingerprint

Data Fusion
Data fusion
K-means
Cluster Analysis
Clustering
kernel
Clustering algorithms
Information Storage and Retrieval
Rayleigh quotient
Clustering Analysis
Laplacian Matrix
Coefficient
Clustering Algorithm
Objective function
Evaluation

ASJC Scopus subject areas

  • Biochemistry
  • Molecular Biology
  • Computational Theory and Mathematics
  • Computer Science Applications
  • Computational Mathematics
  • Statistics and Probability
  • Medicine(all)

Cite this

Yu, S., Liu, X., Tranchevent, L. C., Glänzel, W., Suykens, J. A. K., de Moor, B., & Moreau, Y. (2011). Optimized data fusion for K-means Laplacian clustering. Bioinformatics, 27(1), 118-126. [btq569]. https://doi.org/10.1093/bioinformatics/btq569

Optimized data fusion for K-means Laplacian clustering. / Yu, Shi; Liu, Xinhai; Tranchevent, Léon Charles; Glänzel, W.; Suykens, Johan A K; de Moor, Bart; Moreau, Yves.

In: Bioinformatics, Vol. 27, No. 1, btq569, 01.2011, p. 118-126.

Research output: Contribution to journalArticle

Yu, S, Liu, X, Tranchevent, LC, Glänzel, W, Suykens, JAK, de Moor, B & Moreau, Y 2011, 'Optimized data fusion for K-means Laplacian clustering', Bioinformatics, vol. 27, no. 1, btq569, pp. 118-126. https://doi.org/10.1093/bioinformatics/btq569
Yu S, Liu X, Tranchevent LC, Glänzel W, Suykens JAK, de Moor B et al. Optimized data fusion for K-means Laplacian clustering. Bioinformatics. 2011 Jan;27(1):118-126. btq569. https://doi.org/10.1093/bioinformatics/btq569
Yu, Shi ; Liu, Xinhai ; Tranchevent, Léon Charles ; Glänzel, W. ; Suykens, Johan A K ; de Moor, Bart ; Moreau, Yves. / Optimized data fusion for K-means Laplacian clustering. In: Bioinformatics. 2011 ; Vol. 27, No. 1. pp. 118-126.
@article{51d9424fbf884acf998e613bc8eaa4dd,
title = "Optimized data fusion for K-means Laplacian clustering",
abstract = "Motivation: We propose a novel algorithm to combine multiple kernels and Laplacians for clustering analysis. The new algorithm is formulated on a Rayleigh quotient objective function and is solved as a bi-level alternating minimization procedure. Using the proposed algorithm, the coefficients of kernels and Laplacians can be optimized automatically. Results: Three variants of the algorithm are proposed. The performance is systematically validated on two real-life data fusion applications. The proposed Optimized Kernel Laplacian Clustering (OKLC) algorithms perform significantly better than other methods. Moreover, the coefficients of kernels and Laplacians optimized by OKLC show some correlation with the rank of performance of individual data source. Though in our evaluation the K values are predefined, in practical studies, the optimal cluster number can be consistently estimated from the eigenspectrum of the combined kernel Laplacian matrix.",
author = "Shi Yu and Xinhai Liu and Tranchevent, {L{\'e}on Charles} and W. Gl{\"a}nzel and Suykens, {Johan A K} and {de Moor}, Bart and Yves Moreau",
year = "2011",
month = "1",
doi = "10.1093/bioinformatics/btq569",
language = "English",
volume = "27",
pages = "118--126",
journal = "Bioinformatics",
issn = "1367-4803",
publisher = "Oxford University Press",
number = "1",

}

TY - JOUR

T1 - Optimized data fusion for K-means Laplacian clustering

AU - Yu, Shi

AU - Liu, Xinhai

AU - Tranchevent, Léon Charles

AU - Glänzel, W.

AU - Suykens, Johan A K

AU - de Moor, Bart

AU - Moreau, Yves

PY - 2011/1

Y1 - 2011/1

N2 - Motivation: We propose a novel algorithm to combine multiple kernels and Laplacians for clustering analysis. The new algorithm is formulated on a Rayleigh quotient objective function and is solved as a bi-level alternating minimization procedure. Using the proposed algorithm, the coefficients of kernels and Laplacians can be optimized automatically. Results: Three variants of the algorithm are proposed. The performance is systematically validated on two real-life data fusion applications. The proposed Optimized Kernel Laplacian Clustering (OKLC) algorithms perform significantly better than other methods. Moreover, the coefficients of kernels and Laplacians optimized by OKLC show some correlation with the rank of performance of individual data source. Though in our evaluation the K values are predefined, in practical studies, the optimal cluster number can be consistently estimated from the eigenspectrum of the combined kernel Laplacian matrix.

AB - Motivation: We propose a novel algorithm to combine multiple kernels and Laplacians for clustering analysis. The new algorithm is formulated on a Rayleigh quotient objective function and is solved as a bi-level alternating minimization procedure. Using the proposed algorithm, the coefficients of kernels and Laplacians can be optimized automatically. Results: Three variants of the algorithm are proposed. The performance is systematically validated on two real-life data fusion applications. The proposed Optimized Kernel Laplacian Clustering (OKLC) algorithms perform significantly better than other methods. Moreover, the coefficients of kernels and Laplacians optimized by OKLC show some correlation with the rank of performance of individual data source. Though in our evaluation the K values are predefined, in practical studies, the optimal cluster number can be consistently estimated from the eigenspectrum of the combined kernel Laplacian matrix.

UR - http://www.scopus.com/inward/record.url?scp=78650580591&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78650580591&partnerID=8YFLogxK

U2 - 10.1093/bioinformatics/btq569

DO - 10.1093/bioinformatics/btq569

M3 - Article

VL - 27

SP - 118

EP - 126

JO - Bioinformatics

JF - Bioinformatics

SN - 1367-4803

IS - 1

M1 - btq569

ER -