Java Recurrent Self Organizing Map

The recurrent self organizing map (RSOM) is a self-orgznizing map that has a recurrent layer. This combines elements of the Jordan Neural Network with the Kohonen Self Organizing Map. RSOM's are good at classifying temporal sequences. For example analyzing historical index information and determine if we are in a period of economic growth or decline.

The recurrent self organizing map (RSOM) is modeled in the Encog Workbench looks like this: