It seems they both perform clustering. They both reduce the dimensionality of the input data and classify further inputs based upon their distance/similarity to the center points. These points then update to accommodate the new data.

I am yet to understand how these two methods are different. I suppose it depends on the problem to be solved. How could each be suited to different problems (advantages/disadvantages)?

  • $\begingroup$ What do they have in common except computing distances? In fact, SOMs are often used as a step before k-means, because SOMs do not produce clusters (instead, you run k-means on the mapped data - SOM itself is not a clustering algorithm!). $\endgroup$ – Has QUIT--Anony-Mousse Jun 16 '18 at 17:56

Your Answer

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

Browse other questions tagged or ask your own question.