Cover Image

Hybrid self organizing map for overlapping clusters

The Kohonen self organizing map is an excellent tool in exploratory phase of data mining and pattern recognition. The SOM is a popular tool that maps high dimensional space into a small number of dimensions by placing similar elements close together, forming clusters. Recently researchers found that...

Full description

Saved in:
Bibliographic Details
Main Authors: Md. Sap, Mohd. Noor, Mohebi, Ehsan
Format: Article
Published: Science & Engineering Research Support Center (SERSC) 2008
Subjects:
QA75 Electronic computers. Computer science
Online Access:http://eprints.utm.my/id/eprint/8945/
http://www.sersc.org/journals/IJSIP/vol1_no1/papers/02.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
id my.utm.8945
record_format eprints
spelling my.utm.89452010-10-20T09:31:24Z http://eprints.utm.my/id/eprint/8945/ Hybrid self organizing map for overlapping clusters Md. Sap, Mohd. Noor Mohebi, Ehsan QA75 Electronic computers. Computer science The Kohonen self organizing map is an excellent tool in exploratory phase of data mining and pattern recognition. The SOM is a popular tool that maps high dimensional space into a small number of dimensions by placing similar elements close together, forming clusters. Recently researchers found that to capture the uncertainty involved in cluster analysis, it is not necessary to have crisp boundaries in some clustering operations. In this paper to overcome the uncertainty, a two-level clustering algorithm based on SOM which employs the rough set theory is proposed. The two-level stage Rough SOM (first using SOM to produce the prototypes that are then clustered in the second stage) is found to perform well and more accurate compared with the proposed crisp clustering method (Incremental SOM) and reduces the errors. Science & Engineering Research Support Center (SERSC) 2008 Article PeerReviewed Md. Sap, Mohd. Noor and Mohebi, Ehsan (2008) Hybrid self organizing map for overlapping clusters. International Journal of Signal Processing, Image Processing and Pattern Recognition, 1 (1). pp. 11-20. ISSN 2005-4254 http://www.sersc.org/journals/IJSIP/vol1_no1/papers/02.pdf
institution Universiti Teknologi Malaysia
building UTM Library
collection Institutional Repository
continent Asia
country Malaysia
content_provider Universiti Teknologi Malaysia
content_source UTM Institutional Repository
url_provider http://eprints.utm.my/
topic QA75 Electronic computers. Computer science
spellingShingle QA75 Electronic computers. Computer science
Md. Sap, Mohd. Noor
Mohebi, Ehsan
Hybrid self organizing map for overlapping clusters
description The Kohonen self organizing map is an excellent tool in exploratory phase of data mining and pattern recognition. The SOM is a popular tool that maps high dimensional space into a small number of dimensions by placing similar elements close together, forming clusters. Recently researchers found that to capture the uncertainty involved in cluster analysis, it is not necessary to have crisp boundaries in some clustering operations. In this paper to overcome the uncertainty, a two-level clustering algorithm based on SOM which employs the rough set theory is proposed. The two-level stage Rough SOM (first using SOM to produce the prototypes that are then clustered in the second stage) is found to perform well and more accurate compared with the proposed crisp clustering method (Incremental SOM) and reduces the errors.
format Article
author Md. Sap, Mohd. Noor
Mohebi, Ehsan
author_facet Md. Sap, Mohd. Noor
Mohebi, Ehsan
author_sort Md. Sap, Mohd. Noor
title Hybrid self organizing map for overlapping clusters
title_short Hybrid self organizing map for overlapping clusters
title_full Hybrid self organizing map for overlapping clusters
title_fullStr Hybrid self organizing map for overlapping clusters
title_full_unstemmed Hybrid self organizing map for overlapping clusters
title_sort hybrid self organizing map for overlapping clusters
publisher Science & Engineering Research Support Center (SERSC)
publishDate 2008
url http://eprints.utm.my/id/eprint/8945/
http://www.sersc.org/journals/IJSIP/vol1_no1/papers/02.pdf
_version_ 1643645084878503936
score 13.211869

Similar Items