Could anyone help me how to group the maximum distance data points into clusters.

A is a matrix that contains the location of data points

A=  [8.1     1.7;
     3.9     7.4;
     7.7     3.6;
     9.2     6.7;
     3.8     9.1;
     7.3     9.3]

I have computed the distance

B=pdist(A) 

for each and every data point using euclidean distance formula and using

C=squareform(B)

the following result was obtained.

C=[ 0         7.0803    1.9416    5.1196    8.5586    7.6420
    7.0803         0    5.3740    5.3460    1.7029    3.8949
    1.9416    5.3740         0    3.4438    6.7424    5.7140
    5.1196    5.3460    3.4438         0    5.9093    3.2202
    8.5586    1.7029    6.7424    5.9093         0    3.5057
    7.6420    3.8949    5.7140    3.2202    3.5057         0]

Now i want to group the data points into clusters based on the above computed C matrix under the condition the sum of pairwise distances between the data points in the same cluster should be maximum.

Could anyone please help me on this.

[darova]

Maybe clusterdata? Or kmeans?