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...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Published: |
Science & Engineering Research Support Center (SERSC)
2008
|
| Subjects: |
| _version_ | 1796854465354006528 |
|---|---|
| author | Md. Sap, Mohd. Noor Mohebi, Ehsan |
| author_facet | Md. Sap, Mohd. Noor Mohebi, Ehsan |
| author_sort | Md. Sap, Mohd. Noor |
| collection | ePrints |
| 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. |
| first_indexed | 2024-03-05T18:14:28Z |
| format | Article |
| id | utm.eprints-8945 |
| institution | Universiti Teknologi Malaysia - ePrints |
| last_indexed | 2024-03-05T18:14:28Z |
| publishDate | 2008 |
| publisher | Science & Engineering Research Support Center (SERSC) |
| record_format | dspace |
| spelling | utm.eprints-89452010-10-20T09:31:24Z http://eprints.utm.my/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 |
| spellingShingle | QA75 Electronic computers. Computer science Md. Sap, Mohd. Noor Mohebi, Ehsan Hybrid self organizing map for overlapping clusters |
| title | 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_short | Hybrid self organizing map for overlapping clusters |
| title_sort | hybrid self organizing map for overlapping clusters |
| topic | QA75 Electronic computers. Computer science |
| work_keys_str_mv | AT mdsapmohdnoor hybridselforganizingmapforoverlappingclusters AT mohebiehsan hybridselforganizingmapforoverlappingclusters |