k-means clustering algorithm

4 views (last 30 days)
AS
AS on 8 Jul 2020
I have a data with dimension 1801 by 88, to be clustered into k regions. Intially, all the centres are assigned randomly. the 'i' th data , sig is assigned to the nearest data cluster, c based on minimum Euclidean distance, where i=1:1801 and j=1:k. After completeing the assigning process for all the data, the new position of centre is measured according to
I have written the accordingly, but is not run and I am unable to understand about new position of the centre, although I have written. But not clear to me. How to add number of iteration in this code.
please help me
load signal.dat;
sig=signal(:,1:88);
[maxrow,maxcol]=size(sig);
k=1;
for i=1:k
p(i,:)=randperm(size(sig,1));
end
for i=1:k
for j=1:88
c(i,j)=sig(p(i),j);%random initialization
end
end
temp=zeros(maxrow,1);
while 1
for i1=1:1801
for j=1:88
d(:,j)=sum(dist2(sig(:,j),c(1,j)));
end
[z,g]=max(d,[],1);
if g==temp
break;
else
temp=g;
end
for i=1:k
f=find(g==i);
if f
c(i,:)=mean(sig(find(g==i),:),1);
end
end
end
end

Answers (0)

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!