The SOM has several beneficial features which make it a useful methodology in KDD. It follows the probability density function of the underlying data, it is readily explainable, simple and - perhaps most importantly - highly visual. Most useful it is as a data exploration tool [16, 45]. Depending on the problem, the data mining expert may choose another method for the final analysis, but the strength of the SOM is that it performs many tasks simultaniously equaling the results of other widely used algorithms. The SOM is very effective in clustering and data reduction but it can also be used for data cleaning and preprocessing. Integrated with other methods it can be used for rules extraction [44] and regression [35].