sim_transform

PURPOSE

Transformation of points to compute pairwise similarities of projected data Eqs.(17)-(18)

SYNOPSIS

function out = sim_transform(proj, a, d, weights)

DESCRIPTION

Transformation of points to compute pairwise similarities of projected data Eqs.(17)-(18)
OUT = SIM_TRANSFORM(PROJ, A, D, WEIGHTS)

 Returns:
    (OUT): Points after transformation

 Inputs:
    (PROJ): Projected data
    (A): Range parameter (alpha)
    (D): Delta parameter of transformation function
    (WEIGHTS): Number of original observations associated with each micro-cluster center
        (If micro-clustering has been used)

CROSS-REFERENCE INFORMATION

This function calls:

This function is called by:

f_df_sc Function value and derivative of second smallest eigenvalue of normalised Laplacian
f_sc Second smallest eigenvalue of Normalised Laplacian and difference to 3rd smallest
scppNJW Clustering of projected data using Ng, Jordan and Weiss (2002) normalised spectral clustering algorithm