Acquisition of embodied knowledge on sport skill using TAM network

In this paper, we discuss sport technique evaluation using motion analysis model by neural networks and data mining methods. For students of university, we recorded the continuous forehand stroke of the table tennis in the video frames, and analyzed the trajectory pattern of nine marking points attached at subject's body with a coach's technique evaluation and the motion analysis model. As a result, we obtained some technique rules classified member of table tennis club, middle level player and beginner as fuzzy rules, and also estimated the movement of the marking points to improve in table tennis technique.