Unsupervised Information Extraction and Retrieval

Self-organizing map

Self-organizing maps, also called Kohonen maps, are a kind of neuronal network that are based on unsupervised learning and competition. The neurons are placed on 2-D squared net, although greater dimensions may be used.

The topology of the net is composed of the elements:

• The input layer which propagates the data to the next layer.

• Kohonen map neurons layer, typically a 2-D squared layer. Neurons compete for getting the best output. The winner neuron inhibits another near neurons for strengthening the learning.

The Main application is to group the information of unknown characteristics.

This site is developed for the assignement Information and Retrieval Organization of the subject Information Retrieval and Access over Computer Engineering of Carlos III University of Madrid.

Available versions:

Download the DOC article

Download the PDF article