lyngby_ikm_bic

(export/lyngby/lyngby_ikm_bic.m)

Function Synopsis

bic = lyngby_ikm_bic(X, Centers, Assign, clusters, varargin)

Help text

 lyngby_ikm_bic        - Bayesian information criterion for K-means

       bic = lyngby_ikm_bic(X, Assign, Centers, clusters)

       Input:  X        Data matrix
               Center   Cluster centers
               Assign   Assignment vector matrix
               clusters List of number of clusters

       Output  bic      BIC 

       Bayesian information criterion for iterative K-means.

       Example: 
         load iris.txt
         clusters = 1:50;
         [C,A] = lyngby_ikm_main(iris, 'clusters', clusters);
         bic   = lyngby_ikm_bic(iris, C, A, clusters);
         plot(bic), xlabel('Clusters')
         ylabel('Log likelihood/bic');

       See also LYNGBY, LYNGBY_KM_BIC, LYNGBY_IKM_MAIN,
                LYNGBY_KM_MAIN, LYNGBY_KM_LOGLIKELIHOOD,
                LYNGBY_KM_WITHIN, LYNGBY_IKM_PLOT_W.

 $Id: lyngby_ikm_bic.m,v 1.2 2003/02/05 18:52:52 fnielsen Exp $

Cross-Reference Information

lyngby_km_bic  export/lyngby/lyngby_km_bic.m