A COprocessor for KOhonen's self-organizing map (COKOS)
Heike Speckmann, P. Thole, Wolfgang Rosenstiel · 2005
Presents a system which enables easy and fast computation of Kohonen's self-organizing map (SOM). It is a hardware supported system consisting of different parts. A neural coprocessor (COKOS) is connected to a personal computer (PC) by a special, asynchronous interface. The neural coprocessor works vector-component-parallel and processing-element-serial and speeds up the performance of Kohonen's algorithm of the self-organizing map by magnitude. A user friendly interface (ULTIKOS) supports the preprocessing of the input data and the control and assessment of the learning process.