The unit KOHONEN
provides a selforganizing map (SOM) as it has been described by Teuvo Kohonen
[1, 2]. These SOMs are also known as Kohonen maps, or Kohonen neural networks.
More details on this kind of neural networks can be found in the background
information. The unit KOHONEN contains the class TKohonen which simulates
a Kohonen neural network. Details on the implementation can be found in
the online help pages.
[1] Teuvo Kohonen
Self-Organization and Associative Memory.
Springer-Verlag, Berlin-Heidelberg-New York-Tokio, 1989.
[2]Teuvo Kohonen:
Self-Organizing Maps.
Springer-Verlag, Heidelberg 1995.
| KohMap |  The program shows how to train a Kohonen network using the unit KOHONEN. |
kohmap.zip
(416 kB) |
What's new:
- Release 10.7 [Aug-24, 2020]
- no changes
- Release 10.6 [Mar-09, 2018]
- no changes
- Release 10.5 [Oct-10, 2016]
- no changes
- Release 10.4 [Jun-02, 2015]
- no changes
- Release 10.3 [Oct-06, 2014]
- none
- Release 10.2 [May-30, 2013]
- no changes
- Release 10.1 [Oct-29, 2012]
- no changes
- Release 10.0 [Oct-4, 2011]
- no changes
- Release 9.7 [May-31, 2010]
- no changes
- Release 9.51 [Dec-01, 2008]
- no changes
- Release 9.5 [Oct-31, 2008]
- no changes
- Release 9.0 [May-28, 2007]
- no changes
- Release 8.5 [Feb-18, 2006]
- no changes
- Release 8.2 [Nov-03, 2004]
- no changes
- Release 8.1 [Aug-29, 2004]
- available for Delphi™ 8 for .NET
- Release 8.0 [Apr-15, 2004]
- OnFeedback is now of type TOnPercentDoneEvent (was TKHFeedback)
- Release 7.2 [Mar-23, 2003]
- KOHONEN is now part of MathPack
- Release 7.0 [Sep-15, 2002]
- available for Delphi™ 7.0
- the unit is now CLX compatible
- Release 6.5 [May-28, 2002]
- available for C++Builder™ 6.0
-
Release 6.0 [Aug-06, 2001]
- KOHONEN is now available to the public and included in the SDL Component
Suite 6.0