数据聚类实验报告（附代码）(2)

0.333 0.624 0.051 0.043 0.667 0.459 0.627 0.584 0.557 0.584 0.780 0.957 0.027 0.416 0.051 0.043 0.361 0.416 0.525 0.498 0.471 0.416 0.643 0.710 0.196 0.624 0.051 0.082 0.333 0.251 0.576 0.459 0.667 0.459 0.780 0.957 0.027 0.502 0.051 0.043 0.416 0.251 0.510 0.459 0.416 0.290 0.694 0.749 0.224 0.749 0.153 0.125 0.361 0.290 0.541 0.498 0.667 0.541 0.796 1.000 0.224 0.749 0.102 0.043 0.388 0.376 0.541 0.498 0.557 0.208 0.678 0.749 0.278 0.710 0.086 0.043 0.224 0.208 0.337 0.416 0.529 0.584 0.745 0.918 0.165 0.416 0.067 0.043 0.584 0.502 0.592 0.584 0.416 0.290 0.694 0.749 0.082 0.459 0.086 0.043 0.333 0.125 0.510 0.498 0.557 0.376 0.780 0.710 0.306 0.792 0.118 0.125 0.388 0.333 0.592 0.498 0.918 0.416 0.949 0.831 0.196 0.584 0.086 0.043 0.165 0.169 0.388 0.376 0.835 0.376 0.898 0.710 0.165 0.459 0.086 0.000 0.251 0.290 0.490 0.541 0.804 0.667 0.863 1.000 0.137 0.584 0.102 0.043 0.443 0.416 0.541 0.584 0.584 0.290 0.729 0.749 0.000 0.416 0.016 0.000 0.498 0.376 0.627 0.541 0.388 0.208 0.678 0.792 0.388 1.000 0.086 0.125 0.667 0.459 0.576 0.541

0.584 0.502 0.729 0.918 0.224 0.624 0.067 0.082 0.416 0.290 0.525 0.376 0.945 0.749 0.965 0.875 0.224 0.749 0.086 0.082 0.361 0.208 0.490 0.416 0.471 0.082 0.678 0.584 0.224 0.710 0.086 0.125 0.498 0.333 0.510 0.498 0.361 0.333 0.663 0.792 0.224 0.541 0.118 0.165 0.498 0.333 0.627 0.459 0.557 0.290 0.663 0.710 0.196 0.416 0.102 0.043 0.639 0.416 0.576 0.541 0.804 0.502 0.847 0.710 0.251 0.624 0.086 0.043 0.667 0.416 0.678 0.667 0.498 0.416 0.510 0.710 0.110 0.502 0.102 0.043 0.388 0.251 0.424 0.376 0.804 0.416 0.812 0.624 0.306 0.584 0.086 0.125 0.333 0.169 0.459 0.376 1.000 0.749 0.914 0.792 0.333 0.918 0.067 0.043 0.471 0.290 0.694 0.624 0.557 0.333 0.694 0.584 0.196 0.502 0.035 0.043 0.471 0.584 0.592 0.624 0.945 0.416 0.863 0.918 0.165 0.667 0.067 0.000 0.557 0.125 0.576 0.498 0.584 0.459 0.761 0.710 0.224 0.584 0.086 0.043 0.333 0.208 0.510 0.498 0.722 0.459 0.745 0.831 0.055 0.125 0.051 0.082 0.498 0.416 0.612 0.541 0.722 0.459 0.694 0.918 0.196 0.624 0.102 0.208 0.196 0.125 0.388 0.376 0.694 0.502 0.831 0.918 0.137 0.416 0.067 0.082

0.388 0.416 0.541 0.459 0.667 0.416 0.714 0.918 0.082 0.502 0.067 0.043 0.529 0.376 0.561 0.498 0.612 0.416 0.714 0.792 0.196 0.541 0.067 0.043 0.388 0.333 0.525 0.498 0.443 0.416 0.694 0.710 ];

[n,d] = size(x);

bn=round(n/k*rand);%第一个随机数在前1/K的范围内 nc=[x(bn,:);x(2*bn,:);x(3*bn,:);];%初始聚类中心 %nc=[x(bn,:);x(2*bn,:);x(3*bn,:);x(4*bn,:);];% 4类 [cid,nr,centers] = kmeans(x,k,nc)%调用kmeans函数 for i=1:150 if cid(i)==1,

plot(x(i,1),x(i,2),'r*') % 显示第一类 hold on else if cid(i)==2,

plot(x(i,1),x(i,2),'b*') %显示第二类 hold on else

if cid(i)==3,

plot(x(i,1),x(i,2),'g*') %显示第三类 hold on %else

%if cid(i)==4,

%plot(x(i,1),x(i,2),'k*') %显示第四类 %hold on %end end end end end

strt=['红色*为第一类；蓝色*为第二类；绿色*为第三类；黑色*为第四类' ]; text(-4,-3.6,strt);

%kmeans.m主类

function [cid,nr,centers] = kmeans(x,k,nc) [n,d] = size(x); % 设置cid为分类结果显示矩阵

cid = zeros(1,n); oldcid = ones(1,n);

nr = zeros(1,k); maxgn= 100; iter = 1;

while iter < maxgn %计算每个数据到聚类中心的距离 for i = 1:n

dist = sum((repmat(x(i,:),k,1)-nc).^2,2); [m,ind] = min(dist); % 将当前聚类结果存入cid中 cid(i) = ind; end

for i = 1:k

%找到每一类的所有数据，计算他们的平均值，作为下次计算的聚类中心 ind = find(cid==i); nc(i,:) = mean(x(ind,:)); % 统计每一类的数据个数 nr(i) = length(ind); end

iter = iter + 1; end

maxiter = 2; iter = 1; move = 1;

while iter < maxiter & move ~= 0

move = 0; % 对所有的数据进行再次判断，寻求最佳聚类结果 for i = 1:n

dist = sum((repmat(x(i,:),k,1)-nc).^2,2); r = cid(i); % 将当前数据属于的类给r

dadj = nr./(nr+1).*dist'; % 计算调整后的距离 [m,ind] = min(dadj); % 早到该数据距哪个聚类中心最近 if ind ~= r % 如果不等则聚类中心移动 cid(i) = ind;%将新的聚类结果送给cid

ic = find(cid == ind);%重新计算调整当前类别的聚类中心 nc(ind,:) = mean(x(ic,:)); move = 1; end end

iter = iter+1; end

centers = nc; if move == 0

disp('No points were moved after the initial clustering procedure.') else

disp('Some points were moved after the initial clustering procedure.')

end