mcpp

PURPOSE

Maximum Clusterability Projection Pursuit (MCPP) algorithm

SYNOPSIS

function [optHP, idx, spindex] = mcpp(X, pars, labels, colours)

DESCRIPTION

Maximum Clusterability Projection Pursuit (MCPP) algorithm
[OPTHP, IDX, SPINDEX] = MCPP(X, PARS, LABELS, COLOURS)

 Inputs:
    (X): N-by-D Data matrix
    (PARS): Structure containing all parameters of mddc() algorithm
    (LABELS): True clusters; used only for visualisation (optional)
    (COLOURS): Colormap matrix used only for visualisation (optional)

 Output:
    (OPTHP): Maximum clusterability hyperplane (if more than one initial projection vectors
        are used then the one that maximises the splitting criterion pars.split_index())
    (IDX): Binary cluster assignment {-1,1}
    (SPINDEX): Value of splitting index criterion

CROSS-REFERENCE INFORMATION

This function calls:

fixLabels Enforces cluster labels to be in the range 1:K
ifelse Shorthand for ternary operator: if-then-else
isOctave Determines whether the environment is GNU Octave (returns TRUE) or MATLAB (returns FALSE)
f_df_mc Function value, derivative, and split point along projection vector of variance ratio clusterability projection index
f_mc Variance ratio clusterability and split point along unit-length (v)
mc_v0 Default projection vector for maximum clusterability projection pursuit
mchp Maximum Clusterability Hyperplane (inherits from HP class)

This function is called by:

mcdc Maximum Clusterability Divisive Clustering
mch Maximum Clusterability Hyperplane
mchp Maximum Clusterability Hyperplane (inherits from HP class)