Applying dynamic self organizing maps for identifying changes in data sequences
Rasika Amarasiri, Damminda Alahakoon · Hybrid Artificial Intelligence Systems · 2003
The Growing Self Organizing Map (GSOM), a variant of the Self Organizing Map has been mainly used to cluster and identify relationships in static data in an unsupervised manner. In this paper we discuss about the capabilities of the GSOM in identifying changes in data sequences. To illustrate this we use the analysis of a web server log using the GSOM and highlight the advantages of using the GSOM over traditional web-log analysis methods.