Self-Organising Data Mining

In the article is described the possibility to automate by means of application of self-organisation and other principles more or less the whole data mining process, what we have named self-organising data mining. There are different GMDH-based modelling algorithms implemented - dimensionality reduction, missing value elimination, active neurons, enhanced network synthesis and creation of systems of equations, validation, combining of alternative models - to make knowledge extraction objective, fast and easy-to-use even for large and complex systems.