the distance matrix and the datasets

[lovelyvivi]

Can you provide a distance matrixused in the code? and in case of your datasets,what'a the meaning of the two columns?

[dringel]

The description in the code (cluster_dp_overlap_rd.m) says:

disp('The only input needed is a distance matrix file')
disp('The format of this file should be: ')
disp('Column 1: id of element i')
disp('Column 2: id of element j')
disp('Column 3: weight(i,j)')

So it seems that each line is an edge between 2 nodes (as identified by their ids in columns 1 and 2) where the edge weight is the value in the 3rd column. Not clear to me, however, whether the edge weight is in fact a similarity, or whether it should be a distance given that this is referred to as a distance matrix file. I assume it to be a similarity.

Note that in the code it mentions that best results are achieved on unweighted graphs. Thus, the code here sets all weights to 1 on line 31 of the code:

_%Although there's an extension of weighted network, the reported best result of
%netscience are from unweighted setting.
xx(:,3)=ones(N1,1);

I wonder if the code works reliably if you comment out line 31 to prevent it from overwriting the true weights?

[Bisma1]

what is the input given to "t" and "a"??
I am a naïve user and this thing confused me..