Application: What are they useful for?

      This technology is very useful in few and very special problems. Roughly speaking, these applications are those in which there is a lot of historic data available and nobody knows exactly the structure and parameters that may model such data. In other words, big amounts of data and high uncertainty, regarding the way the data is produced. Examples are stock values (Forecasting), loan risk, local weather, pattern (faces) recognition, and data mining.