%0 Journal Article
%T Selecci¨®n autom¨¢tica de par¨¢metros en LLE
%A Valencia Aguirre
%A Juliana
%A ¨¢lvarez Mesa
%A Andr¨¦s Marino
%A Daza Santacoloma
%A Genaro
%A Acosta Medina
%A Carlos Daniel
%A Castellanos Dom¨ªnguez
%A Germ¨¢n
%J Revista Facultad de Ingenier¨ªa Universidad de Antioquia
%D 2010
%I Scientific Electronic Library Online
%X locally linear embedding (lle) is a nonlinear dimensionality reduction technique, which preserves the local geometry of high dimensional space performing an embedding to low dimensional space. lle algorithm has 3 free parameters that must be set to calculate the embedding: the number of nearest neighbors k, the output space dimensionality m and the regularization parameter a. the last one only is necessary when the value of k is greater than the dimensionality of input space or data are not located in general position, and it plays an important role in the embedding results. in this paper we propose a pair of criteria to find the optimum value for the parameters k and ¦Á, to obtain an embedding that faithfully represent the input data space. our approaches are tested on 2 artificial data sets and 2 real world data sets to verify the effectiveness of the proposed criteria, besides the results are compared against methods found in the state of art.
%K dimensionality reduction
%K locally linear embedding
%K number of nearest neighbors
%K automatic regularization.
%U http://www.scielo.org.co/scielo.php?script=sci_abstract&pid=S0120-62302010000600017&lng=en&nrm=iso&tlng=en