ClusterTools Contents

ClusterTools User Guide

actleval

ACTLEVAL

Evaluate clustering by active learning performance

    ERR = ACTLEVAL(A,PROC)
    ERR = A*ACTLEVAL(PROC)

Input
 A Labeled PRTools dataset
 PROC Cell array of cluster procedure names or mappings, default
 CLUSTF, CLUSTK, CLUSTH, MEANSHIFT, MODECLUSTF

Output
 ERR Error structure (see PLOTE) containing test errors

Description

Evaluation of clustering procedures PROC by constructing an active  learning curve, see MCLUSTLCURVE. Cluster labels are interpreted as  indices of the labeled dataset A. The resulting cluster labels are  compared with the true labels of A. ERR refers to the performance on the  entire dataset and does not distinguish train and test errors.

If no output (E) is requested ACTLEVAL calls PLOTE internally and  thereby produces automatically a figure with a learning curve.

Example(s)

 A = gendatm(3000);
 E = A*actleval({clustm,clustk([],'KCentres')});
 plote(E)

See also

datasets, mappings, knnc, cluste, clusth, clustk, clustkh, clustm, clustf, clustr, dcluste, dclustf, dclusth, dclustk, dclustm, dclustr, clusteval, clustcerr, clustc, clustnum, plote,

ClusterTools Contents

ClusterTools User Guide

This file has been automatically generated. If badly readable, use the help-command in Matlab.