• Dec 21, 2018 News!Vol.11, No. 1 has been published with online version.   [Click]
  • Aug 06, 2018 News! Vol.9, No.1- Vol.9, No.4 has been indexed by EI(Inspec)!   [Click]
  • Sep 30, 2018 News!Vol.10, No. 1-No. 6 has been indexed by Crossref.
General Information
Prof. T. Hikmet Karakoc
Anadolu University, Faculty of Aeronautics and Astronautics, Turkey

IJET 2010 Vol.2(4): 340-344 ISSN: 1793-8236
DOI: 10.7763/IJET.2010.V2.144

Parameter Selection for EM Clustering Using Information Criterion and PDDP

Ujjwal Das Gupta,Vinay Menon and Uday Babbar

Abstract—This paper presents an algorithm to automatically determine the number of clusters in a given input data set, under a mixture of Gaussians assumption. Our algorithm extends the Expectation- Maximization clustering approach by starting with a single cluster assumption for the data, and recursively splitting one of the clusters in order to find a tighter fit. An Information Criterion parameter is used to make a selection between the current and previous model after each split. We build this approach upon prior work done on both the K-Means and Expectation-Maximization algorithms. We extend our algorithm using a cluster splitting approach based on Principal Direction Divisive Partitioning, which improves accuracy and efficiency

Index Terms—clustering, expectation-maximization, mixture of Gaussians, principal direction divisive partitioning

Department of Computer Engineering, Delhi College of Engineering, India (ujjwal.das.gupta@coe.dce.edu, vinay.menon@coe.dce.edu, uday.babbar@coe.dce.edu)


Cite: Ujjwal Das Gupta, Vinay Menon and Uday Babbar, "Parameter Selection for EM Clustering Using Information Criterion and PDDP," International Journal of Engineering and Technology vol. 2, no. 4, pp. 340-344, 2010.

Copyright © 2008-2017. International Journal of Engineering and Technology. All rights reserved. 
E-mail: ijet@vip.163.com