e-DAS


Kohonen Network

A Kohonen Network is a self-organizing neural network trained by an unsupervised learning algorithm. It can be used to classify a set of input vectors according to their similarity. The output of such a network is usually a two-dimensional map, thus, the Kohonen Network is a method for projecting objects from a multidimensional space into a two-dimensional space. This projection keeps the topology of the multi-dimensional space, i.e., points which are near to one another in the multidimensional space are neighbors in the two-dimensional space as well. An advantage of this method is that the results of such a mapping can easily be visualized.

A Kohonen Network has a cubic structure; it has a two-dimensional arrangement of neurons, each having m weights. If the input for the network is a set of m-dimensional vectors the network’s architecture is n x l x m dimensional. The network adapts its values with respect to the input values and thus reflects the input data. This approach is unsupervised learning as the adaptation is merely done with respect to the data describing the individual objects.