10
Self-Organizing Map
This chapter describes the self-organizing map (SOM), which is based on the architecture of artificial neural networks and is used for data clustering and visualization. A list of software packages for SOM is provided along with references for applications.
10.1 Algorithm of Self-Organizing Map
SOM was developed by Kohonen (1982). SOM is an artificial neural network with output nodes arranged in a q-dimensional space, called the output map or graph. The one-, two-, or three-dimensional space or arrangement of output nodes, as shown in Figure 10.1, is usually used so that clusters of data points can be visualized, because similar data points are represented by nodes that are close to each other in the output map. ...
Get Data Mining now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
