A Cluster Compositional Algorithm for Incorporation of Multiple Sets of Clusters of Identical Data
Tahmim Jeba, Tarek Mahmud, Nadia Nahar · 2018
Cluster analysis has become a foremost practice in innumerable sectors. When different techniques produce multiple set of clusters from the same problem domain, these sets demand to be incorporated together to produce a better solution. In this paper, a novel cluster compositional algorithm is proposed to address this issue. The proposed two phased approach intends to assimilate two sets of clusters into one. Firstly, two sets of clusters are compared thoroughly to generate a cumulative set of clusters. Secondly, the approach checks the newly identified set to find out the single element clusters and merges those with other clusters from the set. A case study demonstrates how the approach successfully compares and merges two sets of clusters from a same dataset and results into a more convenient solution.