f_sc
PURPOSE
Second smallest eigenvalue of Normalised Laplacian and difference to 3rd smallest
SYNOPSIS
function [f, eigenGap] = f_sc(v, X, pars)
DESCRIPTION
Second smallest eigenvalue of Normalised Laplacian and difference to 3rd smallest [F, EIGENGAP] = F_SC(V, X, PARS) Inputs: (V): Projection matrix (vector storing matrix column-wise) (X): Data matrix (PARS): parameter struct containing (sigma): scaling parameter for Gaussian kernel (weights): weights of micro-clusters (if not used should be empty) (beta), (delta): parameters of similarity transform function (omega): penalty term used to ensure orthonormality of (v) Output: (F): Second smallest eigenvalue of Normalised Laplacian (EIGENGAP): Difference between 3rd and 2nd smallest eigenvalues
CROSS-REFERENCE INFORMATION
This function calls:- sim_transform Transformation of points to compute pairwise similarities of projected data Eqs.(17)-(18)
- schp Class implementing a linear projection subspace of arbitrary dimensions estimated through SCPP
- scpp Spectral Clustering Projection Pursuit (SCPP) (divisive clustering is implemented in scppdc.m)
Generated on Tue 17-Jul-2018 18:58:09 by m2html © 2005