Recognisable Shapes for Self-Organizing Maps

The Self-Organizing Map (SOM), and other related architectures, enjoy a growing popularity in the field of Data Mining. These neural network algorithms provide a topology-preserving mapping from high-dimensional data to a lower dimension, which allows for an easier interpretation of complex data. For visualisation of trained maps, a lot of different techniques have been developed. However, convenient and practical methods for describing the (normally) rectangular maps have not yet been subject of intensive research, and are still missing. In this paper, new shapes for self organising architectures, which allows for an easier explanation, will be presented. These map layouts are oriented on shapes well-known to readers, as for example country or continent maps, or geometrical shapes.