f_mc

PURPOSE

Variance ratio clusterability and split point along unit-length (v)

SYNOPSIS

function [f, bmin] = f_mc(v,X,minsize)

DESCRIPTION

Variance ratio clusterability and split point along unit-length (v)
[F, BMIN] = F_MC(V,X,MINSIZE)

 Inputs:
    (V): Projection vector
    (X): N-by-D data matrix
    (MINSIZE): Minimum cluster size

 Output:
    (F): Variance Ratio Clusterability of projected dataset X*(V./norm(V,2))
    (BMIN): Optimal split point along unit-length vector (V./norm(V,2))

CROSS-REFERENCE INFORMATION

This function calls:

This function is called by:

mc_spindex Default split_index used to select which cluster MCDC partitions at each iteration
mchp Maximum Clusterability Hyperplane (inherits from HP class)
mcpp Maximum Clusterability Projection Pursuit (MCPP) algorithm