SOM(self-organizing map)

Question

hello

How to use SOM as a supervised learning technique? which approach can be added to SOM to turn it to supervised learning?

Answer 1

This is an old question, but SOM is not a supervised learning approach. But if can be used to reduce the dimensionality of your data to use for a supervised learning algorithm.

For example, SOM may be used to group long input sequences together based on features that it discovers into vectors of a much smaller size--say, using SOM to reduce 100,000 1000-length sequences to 100,000 (or fewer) 100-length sequences. These resultant vectors may then be used to train a supervised algorithm.