Title: A new improved cluster validity indexing technique: harnessed from Goodman-Kruskal validity index

Authors: Smita Prava Mishra; Debahuti Mishra; Srikanta Patnaik

Addresses: School of Computer Science and Engineering, Institute of Technical Education and Research, Siksha O Anusandhan University, Bhubaneswar, Odisha, India ' School of Computer Science and Engineering, Institute of Technical Education and Research, Siksha O Anusandhan University, Bhubaneswar, Odisha, India ' School of Computer Science and Engineering, Institute of Technical Education and Research, Siksha O Anusandhan University, Bhubaneswar, Odisha, India

Abstract: The true potential of clustering techniques is not harnessed optimally because of several reasons. Clustering is implemented either on the preclassified datasets or if implemented on unclassified datasets, it remains unacceptable because its validity cannot be established. Cluster validity techniques come to rescue in the latter cases. Several internal and external cluster validity indices are studied and used to validate the clustering techniques. Moreover, the validity of the indexing techniques is needed to be established first. The current work suggests an improvement over the Goodman-Kruskal indexing technique and establishes its validity by applying it on several benchmark datasets. Hence, it suggests a new cluster validity indexing technique.

Keywords: cluster validity indexing; indexing techniques; Goodman-Kruskal validity index; clustering.

DOI: 10.1504/IJICT.2015.065994

International Journal of Information and Communication Technology, 2015 Vol.7 No.1, pp.88 - 99

Received: 12 Oct 2013
Accepted: 03 Jan 2014

Published online: 17 Nov 2014 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article