A Multiscale Clustering Approach for Non-IID Nominal Data

Algorithm 2 Upscaling algorithm based on CMS (UACMS).

 Input: list (Rcenter [1]11, ..., Rcenter [1]1k,…, Rcenter[m]mk)//the clustering result of benchmark-scale dataset

 Output: the clustering result of target dataset

 1: simlist = NULL, cs = list(Rcenter [1]11, ..., Rcenter [1]1k,…, Rcenter[m]mk)

 2: for i in range(m)://m is the number of partition

 3:  for j in range(ik)://ik is the number of cluster belonged ith partition

 4:   computing similarity(A, B) according to (4)

 5: do{

 6:  choose minimum from simlist

 7:  merge two clusters

 8:  Update the cs

 9: }while (condition is true)

 10: return cs//the clustering result of target dataset