CciMST: A Clustering Algorithm Based on Minimum Spanning Tree and Cluster Centers

Algorithm 2

Pseudocode of Intercluster separation.

) Input: Data set , the total number of points n, the number of clusters K,

the clustering result , K-1 inconsistent edges , the

geodesic distance between points x_i and x_j

) Output: Intercluster separation Sep

) Begin

Construct K-1 pairs of adjacent clusters according to ( The two end points

of e_i belong to Cluster_i and Cluster_j.)

Calculate the intercluster distance between adjacent clusters and

(5.1) Select a pair of adjacent clusters

(5.2) Calculate the minimum geodesic distance from each point in

the to all of the points in the

(5.3) Sort all of the minimum geodesic distances in ascending order

(5.4) Sum up the top 20% minimum geodesic distances (Here, suppose there are a

total of minimum geodesic distances)

(5.5) Similar to Step (5.4), for the adjacent clusters , sum up the top 20% minimum

geodesic distances . (Here, suppose there are a total of minimum geodesic

distances)

(5.6) Calculate the distance between and

Calculate the average of the K-1

) End