Given a converged Kohonen feature map, how would one evaluate the clustering in terms of intra- and inter-cluster distances?
Assuming that both the trained codebook vectors and Unified Distance Matrix are available. Currently I have been evaluating them empirically using heatmaps of the UMatrix. However, I was wondering if there was a more reliable, statistical method.
You could use a relative small map and consider each node a cluster, but this is far from optimal. If you want to apply an automated cluster detection method you should definitely read
Clustering of the Self−Organizing Map
and search similar bibliography.
You could also use more sophisticated versions of SOM algorithm (multi leveled, self growing, etc).
In any case, keep in mind that the problem of finding the "correct" number of clusters doesn't have a finite solution.
Meta info: Basically I am copying my answer for a similar question in stack overflow. Probably not the proper way to do it, but how else?
