Please use this identifier to cite or link to this item:
|Type:||Artigo de evento|
|Title:||Automatic Data Classification By A Hierarchy Of Self-organizing Maps|
|Author:||Costa Jose Alfredo F.|
de Andrade Netto Marcio L.
|Abstract:||Clustering is the process by which discrete objects are assigned to groups that have similar characteristics. The self-organizing maps (SOM) have been widely used as a data visualization tool. Some of its advantages include information compression and density estimation while trying to preserve topological and metric relationship of the primary data items. For using SOM as a clustering tool it is required additional procedures to interpret the mapping obtained through unsupervised learning. Costa & Netto (1999) described the usage of image analysis and mathematical morphology to find automatically regions of similar neurons and their borders. The purpose of this paper is to enhance the clustering process in order to detail the underlying structure obtained on a first trial. Groups of neurons associated to clusters are further subdivided in new sub-networks, generating a tree-like structure of SOMs. Differently of other hierarchical SOM approaches, the number of sub-nets for a given SOM in a given height of the tree is not specified in advance. The process can be seen as a dynamic strategy for cluster discovery.|
|Editor:||IEEE, United States|
|Appears in Collections:||Unicamp - Artigos e Outros Documentos|
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.