SOM with False-Neighbor degree and its behaviors

This study improves SOM with False-Neighbor degree between neurons (FN-SOM) which solves some problems of the original FN-SOM. The improved FN-SOM changes the neighborhood relationship more flexibly according to the situation and the shape of data. We investigate the behavior of FN-SOM in detail. Learning performance is evaluated both visually and quantitatively using the three measurements. We confirm effectively of the improved FN-SOM with comparing with the conventional SOM, Growing Grid and the original FN-SOM.