cluster_stat.m

動態聚類k-means演算 將輸入在程式中的數據資料 給予適當的分群

function [dbelong, neach, avedist, quanterr] = ...
    cluster_stat(data, clusters, nc,cl,class);

% function [dbelong, neach, avedist, quaterr] = 
%          kmeans_stat(data, centers, nc)
%

% Compute the distance between each datapoint to the clusters
dists = kmeans_dist(data, clusters);

% Assign datapoints to clusters and compute updates
[minval, dbelong] = min(dists');

% How many in each cluster, avedist and quanterr
quanterr = 0.0;
for i=1:nc
  idxtmp = find(dbelong == i & class == cl);
  neach(i) = length(idxtmp);
  
  if length(idxtmp) > 0
    cpoints = data(idxtmp,:);
    ctmp = clusters(i,:);
    cdists = kmeans_dist(cpoints, ctmp);
    avedist(i) = mean(cdists);
    quanterr = quanterr + sum(cdists);
  else
    avedist(i) = -1.0;
  end
  
end