Towards data-based assessment of individual tactics skills in team sports based on fuzzy petri nets
|Titel||Towards data-based assessment of individual tactics skills in team sports based on fuzzy petri nets|
|Organisation||Software Competence Center Hagenberg|
This paper addresses the problem of designing an explanatory computational model for the assessment of individual tactic skills in team sports. The modeling approach tackles the complexity and difficulty of this problem by fusing fuzzy human-like knowledge related to tactical behavior with time-continuous position data from a tracking system. For this purpose a four layer based hierarchical conceptional architecture is proposed. The bottom layer is represented by physically meaningful variables derived from time- continuous position data. Based thereupon, we introduced a temporal segmentation layer that relates the physical variables to game-situation specific temporal phases. Conceptually this layer can be represented by a Fuzzy Petri Net which plays the role of the interface between the trainer's expert knowledge in terms of fuzzy rules about the temporal phases and the time-continuous position data. In the next layer, unsupervised clustering techniques are applied in the feature space induced by the states of the Fuzzy Petri Net. Finally, the resulting clusters of the middle layer are interpreted in terms of key performance indicators in the top layer in order to provide a meaningful explanatory model for the assessment. This approach is illustrated and examined statistically by means of one-versus-one situations in association football.