Method and apparatus for generating a not parametric, data based functional model

摘要

The present invention relates to a method for determining an not parametric, data based functional model with training data, wherein the training data provided a number of input variables defined by one or more measuring points with respective associated output values of an output variable, comprising the steps of:– Providing (s) of different uncertainty values for a part of the or all measuring points of the training data; and– Determining (s4) of the not parametric, data based functional model according to an algorithm of the of the measuring points of the input variables, the respective associated output values of the output variable and depends on the respectively associated uncertainty values.