Multiway Spectral Clustering with Out-of-Sample Extensions through Weighted Kernel PCA

Authors:
Carlos Alzate;Johan A. K. Suykens
Affiliations:
Katholieke Universiteit Leuven, Leuven;Katholieke Universiteit Leuven, Leuven
Venue:
IEEE Transactions on Pattern Analysis and Machine Intelligence
Year:
2010

Citing 0
Cited 11

Identifying Customer Profiles in Power Load Time Series Using Spectral Clustering

ICANN '09 Proceedings of the 19th International Conference on Artificial Neural Networks: Part II
A regularized formulation for spectral clustering with pairwise constraints

IJCNN'09 Proceedings of the 2009 international joint conference on Neural Networks
Sparse kernel spectral clustering models for large-scale data analysis

Neurocomputing
A new anticorrelation-based spectral clustering formulation

ACIVS'11 Proceedings of the 13th international conference on Advanced concepts for intelligent vision systems
Sparse spectral clustering method based on the incomplete Cholesky decomposition

Journal of Computational and Applied Mathematics
Hierarchical kernel spectral clustering

Neural Networks
Constrained spectral embedding for K-way data clustering

Pattern Recognition Letters
Risk group detection and survival function estimation for interval coded survival methods

Neurocomputing
Localized matrix factorization for recommendation based on matrix block diagonal forms

Proceedings of the 22nd international conference on World Wide Web
Local information-based fast approximate spectral clustering

Pattern Recognition Letters
Real-time fault diagnosis for gas turbine generator systems using extreme learning machine

Neurocomputing

Quantified Score

Hi-index	0.14

Visualization

Abstract

A new formulation for multiway spectral clustering is proposed. This method corresponds to a weighted kernel principal component analysis (PCA) approach based on primal-dual least-squares support vector machine (LS-SVM) formulations. The formulation allows the extension to out-of-sample points. In this way, the proposed clustering model can be trained, validated, and tested. The clustering information is contained on the eigendecomposition of a modified similarity matrix derived from the data. This eigenvalue problem corresponds to the dual solution of a primal optimization problem formulated in a high-dimensional feature space. A model selection criterion called the Balanced Line Fit (BLF) is also proposed. This criterion is based on the out-of-sample extension and exploits the structure of the eigenvectors and the corresponding projections when the clusters are well formed. The BLF criterion can be used to obtain clustering parameters in a learning framework. Experimental results with difficult toy problems and image segmentation show improved performance in terms of generalization to new samples and computation times.