I am implementing clustering using BFO with dataset. However I have matlab code for it. But I want to find the cost function in java specially the parameters of J.
The code is attached. X is the dataset. Suggest how to find the Cost function:
======================
ww=ones(150,3);
for m=1:s % the initital posistions
wh=randint(1,1,[1 150]);
P(1,m,1,1,1)= X(wh,1);
P(2,m,1,1,1)=X(wh,2);
P(3,m,1,1,1)=X(wh,3);
P(4,m,1,1,1)=X(wh,4);
end
for ell=1:Ned
%Reprodution loop
for K=1:Nre
% swim/tumble(chemotaxis)loop
for j=1:Nc
for i=1:s
J(i,j,K,ell)=Live_fnn(ww,X,P(:,i,j,K,ell)); %COST FUNCTION
% Tumble
Jlast=J(i,j,K,ell);
Delta(:,i)=(2*round(rand(p,1))-1).*rand(p,1);
P(:,i,j+1,K,ell)=P(:,i,j,K,ell)+c(i,K)*Delta(:,i)/sqrt(Delta(:,i)'*Delta(:,i));
Dear Lokesh Gagnani,
What exactly do you mean by "cost function " ? In my opinion, if I understand right the question, it can mean to minimize the sum of squared Euclidean distance between each object and the center of the cluster to which it belongs. Also you can use DB or CS index as cost function in clustering problem.
Best Regards...
I mean to say here the J(i,j,K,ell) has 4 parameters how can write this in JAVA. Further objective function is shown in image and is similar to the euclidean distance.
Thanks
Dear Lokesh Gagnani ,
There is a product called MATLAB Builder JA which will create Java code that can execute the MATLAB code. This product allows you to use the MATLAB functions from Java program.
http://www.mathworks.com/products/javabuilder/description2.html
http://www.mathworks.com/support/2014a/javabuilder/2.3.1/demos/using-builder-ja.html
- Badges
- Science topic
- Similar topics
- Bioinformatics
- Bioinformatics and Computational Biology