K-means MATLAB.function y=kMeansCluster(m,k,isRand)%%%%%%%%%%%%%%%%%% kMeansCluster - Simple k means clustering algorithm % Author: Kardi Teknomo, Ph.D. % % Purpose: classify the objects in data matrix based on the attributes% Criteria: minimize Euclidean distance between centroids and object points % For more explanation of the algorithm, see/kardi/tutorial/kMean/index.html% Output: matrix data plus an additional column represent the group of each object % % Example: m = [ 1 1; 2 1; 4 3; 5 4] or in a nice form% m = [ 1 1; % 2 1; % 4 3; % 5 4] % k = 2 % kMeansCluster(m,k) produces m = [ 1 1 1;% 2 1 1; % 4 3 2; % 5 4 2] % Input:% m - required, matrix data: objects in rows and attributes in columns % k - optional, number of groups (default = 1)% isRand - optional, if using random initialization isRand=1, otherwise input anynumber (default)% it will assign the first k data as initial centroids%% Local Variables% f - row number of data that belong to group i% c - centroid coordinate size (1:k, 1:maxCol)% g - current iteration group matrix size (1:maxRow)% i - scalar iterator% maxCol - scalar number of rows in the data matrix m = number of attributes% maxRow - scalar number of columns in the data matrix m = number of objects% temp - previous iteration group matrix size (1:maxRow)% z - minimum value (not needed) %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%if nargin<3, isRand=0; endif nargin<2, k=1; end[maxRow, maxCol]=size(m)if maxRow<=k,y=[m, 1:maxRow]else% initial value of centroidif isRand,p = randperm(size(m,1)); % random initializationfor i=1:kc(i,:)=m(p(i),:)endelsefor i=1:kc(i,:)=m(i,:) % sequential initializationendendtemp=zeros(maxRow,1); % initialize as zero vectorwhile 1,d=DistMatrix(m,c); % calculate objcets-centroid distances[z,g]=min(d,[],2); % find group matrix gif g==temp,break; % stop the iterationelsetemp=g; % copy group matrix to temporary variable endfor i=1:kf=find(g==i);if f % only compute centroid if f is not empty c(i,:)=mean(m(find(g==i),:),1);endendendy=[m,g];endThe Matlab function kMeansCluster above call function DistMatrix as shown in the code below. It works for multi-dimensional Euclidean distance. Learn about other type of distance here.function d=DistMatrix(A,B)%%%%%%%%%%%%%%%%%%%%%%%%%% DISTMATRIX return distance matrix between points in A=[x1 y1 (1)and in B=[x2 y2 (2)% Copyright (c) 2005 by Kardi Teknomo,/kardi/%% Numbers of rows (represent points) in A and B are not necessarily the same.% It can be use for distance-in-a-slice (Spacing) ordistance-between-slice (Headway),%% A and B must contain the same number of columns (represent variables of n dimensions),% first column is the X coordinates, second column is the Y coordinates, and so on.% The distance matrix is distance between points in A as rows% and points in B as columns.% example: Spacing= dist(A,A)% Headway = dist(A,B), with hA ~= hB or hA=hB% A=[1 2 3; 4 5 6; 2 4 6; 1 2 3]; B=[4 5 1; 6 2 0]% dist(A,B)= [ 4.69 5.83;% 5.00 7.00;% 5.48 7.48;% 4.69 5.83]%% dist(B,A)= [ 4.69 5.00 5.48 4.69;% 5.83 7.00 7.48 5.83]%%%%%%%%%%%%%%%%%%%%%%%%%%%[hA,wA]=size(A);[hB,wB]=size(B);if wA ~= wB, error(' second dimension of A and B must be the same'); endfor k=1:wAC{k}= repmat(A(:,k),1,hB);D{k}= repmat(B(:,k),1,hA);endS=zeros(hA,hB);for k=1:wAS=S+(C{k}-D{k}').^2;endd=sqrt(S);%这是一个简单的k均值聚类批处理函数%待分类的样本x=mvnrnd（mu，siguma，20）%idx3=kmeans(x,3,'distance','city');或者%idx4=kmeans(x,4,'dist','city','display','iter');这个可以显示出每次迭代的距离和%显示分类轮廓图[silh4,h]=silhouette(x,idx4,'city');xlable('silhouette % value');ylab le('cluster')%mean(silh5) 结果越接近1越好mu1=[1,1];sigma1=[0.5 0;0 0.5];x1=mvnrnd(mu1,sigma1,10);mu2=[7,7];sigma2=[0.5 0;0 0.5];x2=mvnrnd(mu2,sigma2,10);x=[x1;x2]plot(x(:,1),x(:,2),'bo');[idx2,c]=kmeans(x,2,'dist','city','display','iter');figure(2);[silh2,h]=silhouette(x,idx2,'city');%xlable('silhouette value')%ylable('cluster')figure(3);plot(x(idx2==1,1),x(idx2==1,2),'r+',x(idx2==2,1),x(idx2==2,2),'b.'); '分类水平：（1为最好）：'a=mean(silh2);a'图心矩阵为：'c。
