Algorithm 1. Cluster Setup Algorithm

Inputs: (1) n objects

(2) k—number of clusters

Result:k clusters of different sizes

1: Each object constitutes an initial cluster;

2: Current cluster number k’ = n;

3: while (k’ > k)

4:   for i = 1 → k’

5:     for j = 1 → k’

6:       if i ~= j

7:         calculate the distance of the two nodes in the two clusters Ci and Cj;

8:         build a distance set D and obtain the distance mean Davg and the variance δ2;

9:         calculate the clustering factor of clusters Ci and Cj $$F_{i,j}(D_{avg},{\delta }^2)=\frac{1}{D_{avg}+0.5{\delta }^2}$$

10:     end if

11:   end for

12: end for

13: combine the two clusters corresponding to Fmax

14: k’--;

15: end while