A new method for semi-automatic fuzzy training and its application in environmental modeling

A new method for semi-automatic fuzzy training is introduced. This technique combines statistical data analysis with the training of fuzzy rules. The modeler controls the fuzzy training system with easily-understood parameters. An incomplete set of fuzzy rules can be refined by the expert as well as reveal information about the data set. The effectiveness of the new method is demonstrated using two examples. The fuzzy training algorithm was implemented as open source software with a graphical user interface.

► A new method for semi-automatic fuzzy training is introduced.
► The modeler controls the fuzzy training system with easily-understood parameters.
► The effectiveness of the new method is demonstrated using two examples.
► The fuzzy training algorithm was implemented as open source software.