Self-organizing maps

The SOM is an algorithm used to visualize and interpret large high-dimensional data sets. Typical applications are visualization of process states or financial results by representing the central dependencies within the data on the map.

The research activities of the group include applications of the Self-Organizing Map (SOM) and Learning Vector Quantizing (LVQ) algorithms of Academician Kohonen.

Some background material of the SOM

• Short description of the SOM algorithm
• Book: Self-Organizing Maps (Kohonen, 1995, 1997, 2001)
• Book: Kohonen Maps, (Eds. Oja & Kaski, 1999)
• WSOM'97 - Workshop on Self-Organizing Maps, Helsinki University of Technology, Finland, June 4-6, 1997
• WSOM'01 - Workshop on Self-Organizing Maps, Lincoln, England, June 13-15, 2001
• Bibliography on the SOM and the LVQ. Searches on the bibliography can be done on a search server.
• Public domain software: Implementations of the SOM and the LVQ (SOM_PAK and LVQ_PAK), and SOM Toolbox for Matlab
• A theoretical highlight: The Adaptive-Subspace SOM (ASSOM) creates invariant-feature detectors (assom_nc.ps or assom_nc.ps.gz).

Applications

The Self-Organizing Maps have been used at the Research Centre in such applications as:

• Automatic speech recognition
• Clinical voice analysis
• Monitoring of the condition of industrial plants and processes
• Cloud classification from satellite images
• Analysis of electrical signals from the brain
• Organization of and retrieval from large document collections (the WEBSOM method)
• Analysis and visualization of large collections of statistical data (e.g. macroeconomic data)