11
$\begingroup$

After training a self-organising map, one can calculate the U-Matrix. There are some tools to manually visualize it and identify clusters, but I'm wondering if there is any algorithm to do this process in an automatic way (i.e. not having a human looking at the figure to identify the clusters).

Is there any way to do this? I'm writing my code in R. I haven't found anything of the kind on the Internet, so maybe someone can help me out here.

$\endgroup$
11
$\begingroup$

Yes, I've seen clustering algorithms being run on a SOM's U-Matrix. It isn't much common however because SOMs are used to visualize high-dimensional data on 2 dimensions.

An example you can look at this, where a DBSCAN is run on a U-matrix to cluster its neurons. This example is from the Python library 'somoclu'. Unfortunately, I'm not aware of any packages on R.

$\endgroup$
  • $\begingroup$ Thanks for the answer. I was actually looking for some algorithm regarding image segmentation using the U Matriz, and not the nodes of the map themselves. $\endgroup$ – Tendero Oct 2 '18 at 2:29

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.