Self-Organizing Map Demo

To view this content, you need to install Java from java.com

A Self-Organizing Map is a system for unsupervised learning and categorization developed by Helsinki University of Technology professor Teuvo Kohonen in the early 1980s. It uses a mapping of high-dimensional inputs onto a map of units in a way that preserves relative distances between data points. The map units are usually organized in a two-dimensional matrix, which allows easy visualization by mapping the units directly to points on the screen. The Self-Organizing Map is used in visualizations of high-dimensional data because of its clustering abilities.