The author has developed a novel approach to fuzzy modeling from input
-output data, Using the basic techniques of soft computing, the method
allows supervised approximation of multi-input multi-output (MIMO) sy
stems. Typically, a small number of rules are produced, The learning c
apacity of FuGe-NeSys is considerable, as is shown by the numerous app
lications developed. The paper gives a significant example of how the
fuzzy models developed are generally better than those to be found in
literature as concerns simplicity and both approximation and classific
ation capabilities.