ePrints@IIScePrints@IISc Home | About | Browse | Latest Additions | Advanced Search | Contact | Help

Some initialization problems in clustering and their remedies

Ghosh, D and Shivaprasad, AP (1998) Some initialization problems in clustering and their remedies. In: International Conference on Software in Telecommunications and Computer Networks: SoftCOM '98, 14-17 Oct. 1998, Split, Croatia, pp. 595-604.

Full text not available from this repository. (Request a copy)

Abstract

In this paper, we propose a new algorithm for clustering a set of given feature vectors. The algorithm provides a scheme for transition from fuzzy clustering mode to crisp decision based clustering mode. This allows for better estimates of the cluster centers while reducing the dependence of the final cluster centers on the initial choice. The algorithm is also capable of escaping from any local minimum via appropriate partitioning of the input data-set. Thus, the algorithm always leads to the global minimum. Simulation results reveal the merit of this algorithm over all the existing algorithms

Item Type: Conference Paper
Additional Information: Copyright of this article belongs to University fo Split, Croatia.
Department/Centre: Division of Electrical Sciences > Electrical Communication Engineering
Date Deposited: 29 May 2007
Last Modified: 27 Aug 2008 12:40
URI: http://eprints.iisc.ernet.in/id/eprint/10079

Actions (login required)

View Item View Item