Thursday, November 10, 2011

Self-organizing Maps — PyMVPA v2.0.0~rc5 documentation

Self-organizing Maps — PyMVPA v2.0.0~rc5 documentation: "This is a demonstration of how a self-organizing map (SOM), also known as a Kohonen network, can be used to map high-dimensional data into a two-dimensional representation. For the sake of an easy visualization ‘high-dimensional’ in this case is 3D.

In general, SOMs might be useful for visualizing high-dimensional data in terms of its similarity structure. Especially large SOMs (i.e. with large number of Kohonen units) are known to perform mappings that preserve the topology of the original data, i.e. neighboring data points in input space will also be represented in adjacent locations on the SOM."

'via Blog this'