Artificial neural network models of sports motions
Describing and analysing phenomena in sports is necessarily connected with the process of modelling. Modelling starts with the description of the phenomena and it is accompanied by the active and subjective decision of researchers for a limited number of its criteria. Thus, a reduced or simplified mapping of the real object is achieved. Accordingly, the description of a phenomenon means mapping the model original (real object) to a (simplified) model and is part of the process of scientific modelling. In this process of scientific modelling the determination of the model purpose is followed by the selection of adequate model variables that serve as in-and output variables for different forms of models. The criteria for the selection of these variables are manifold and reach from ‘just measurable’ up to highly sophisticated mathematical extraction procedures.