%0 Journal Article
%T Kernel-based framework for spectral dimensionality reduction and clustering formulation: A theoretical study
%A A. E. CASTRO OSPINA
%A D. H. PELUFFO ORD¨®£¿EZ
%A D. VIVEROS MELO
%A M. A. BECERRA
%A M. ORTEGA ADARME
%A Xiomara Patricia BLANCO VALENCIA
%J -
%D 2017
%R http://dx.doi.org/10.14201/ADCAIJ2017613140
%X This work outlines a unified formulation to represent spectral approaches for both dimensionality reduction and clustering. Proposed formulation starts with a generic latent variable model in terms of the projected input data matrix. Particularly, such a projection maps data onto a unknown high-dimensional space. Regarding this model, a generalized optimization problem is stated using quadratic formulations and a least-squares support vector machine. The solution of the optimization is addressed through a primal-dual scheme. Once latent variables and parameters are determined, the resultant model outputs a versatile projected matrix able to represent data in a low-dimensional space, as well as to provide information about clusters. Particularly, proposed formulation yields solutions for kernel spectral clustering and weighted-kernel principal component analysis
%K cluster
%K dimension reduction
%K support vector machine
%K low-dimensional space
%U http://campus.usal.es/~revistas_trabajo/index.php/2255-2863/article/view/ADCAIJ2017613140