Model Based Clustering Using Evolutionary Algorithm

Clustering is collection of data objects that are similar to one another and thus can be treated collectively as one group. The model based clustering approach uses model for clustering and optimizes the fit between the data and model. The evolutionary algorithm has the ability to thoroughly search...

Full description

Saved in:
Bibliographic Details
Published inI-Manager's Journal on Information Technology Vol. 2; no. 4; pp. 16 - 20
Main Authors Radhika, R. A., Priya, D. R.
Format Journal Article
LanguageEnglish
Published Nagercoil iManager Publications 15.11.2013
Subjects
Cluster analysis
Genetic algorithms
Information technology
Online AccessGet full text

Cover

More Information
Summary:Clustering is collection of data objects that are similar to one another and thus can be treated collectively as one group. The model based clustering approach uses model for clustering and optimizes the fit between the data and model. The evolutionary algorithm has the ability to thoroughly search the parameter space, providing an approach inherently more robust with respect to local maxima. In EvolvExpectation-Maximization(EvolvEM)algorithm,Expectation Maximization and Genetic algorithm is used for clustering data which shows more efficiency then EM clustering. The drawback in this method is that its execution time is higher and it requires more parameters. In the proposed approach, instead of Genetic algorithm, Bee colony optimization can be combined with Expectation Maximization algorithm in order to improve execution time and clustering efficiency. Hence, it can be efficiently used for clustering.
ISSN:2277-5110
2277-5250
DOI:10.26634/jit.2.4.2540