How do i know which inputs are clustered together after training using SOM in matlab?

Question

I am new to Self-Organizing Map (SOM) and had been learning and testing it out using Matlab.

I would like to know how do i retrieve inputs that are clustered (similar) together using matlab?

Answer 1

The SOM is a non-supervised clustering algorithm, in which similar samples, are represented closer on the feature map (this is, similar samples will fire nodes that are closer together).

Further, you can try to identify similar regions on the SOM feature map through for example the U-Matrix (map representing the average distance between the node's weight vector and its closest neighbours), the Hit-Map (map representing the number of times the node was chosen as the best matching unit - BMU for the training data), or even each training component.

For a better understanding and an example in matlab see the following question: Dimensionality Reduction using Self Organizing Maps