How do I classify this data?

I am trying to come up with a useful machine learning algorithm, and, having reviewed a lot of code that seems useless to myself, I thought I would do some initial validation criteria based on simple averages and minima/maxima.

The next question is how to classify the information in a function that seems to work well for my data, since I am still in the teething phases of my learning curve.

function ExampleKMeans

n = 100;
r1 = repmat([1,1],n,1) + 2.*randn(n,2);
r2 = repmat([1,10],n,1) + 2.*randn(n,2);
r3 = repmat([10,10],n,1) + 2.*randn(n,2);
%r4 = repmat([10,1],n,1) + 2.*randn(n,2);

data = [r1;r2;r3];

mn = min(data);
mx = max(data);

figure;
subplot(3,4,1);
plot(data(:,1),data(:,2),'r+');
set(gca,'xlim',[mn(1) mx(1)],'ylim',[mn(2),mx(2)]);
title('Original Data');

for k = 2:10
    idx = kmeans(data,k);
    validity = DunnIndex(data,idx,k);

      subplot(3,4,k);
      gscatter(data(:,1),data(:,2),idx);
      set(gca,'xlim',[mn(1) mx(1)],'ylim',[mn(2),mx(2)]);
      title(['Dunns Validity Measure: ',num2str(validity)]);
  end

function validity = DunnIndex(data,idx,K)

N = size(data,1);
totalsum = 0;
for i = 1:K
    x = data(idx == i,:);
    zi = mean(x);

      for j = 1:size(x,1)
          totalsum = totalsum + sum((x(j,:)-zi).^2);        
      end    
  end

intra = totalsum/N;

distmatrix = zeros(K,K);
for i = 1:K-1
    xi = data(idx == i,:);
    zi = mean(xi);

      for j = i+1:K
          xj = data(idx == j,:);
          zj = mean(xj);        
          distmatrix(i,j) = sum((zi-zj).^2);
      end
  end

inter = min(distmatrix(find(distmatrix)));

validity = intra/inter;

function scaledData = minmaxscaling(data)

mn = min(data);
mx = max(data);

for i = 1:size(data,1)
    scaledData(i,:) = (data(i,:) - mn)./(mx-mn);
end