Identifying talent in sepak takraw via anthropometry indexes

This chapter evaluates the importance of different anthropometric indexes towards the categorisation of the ability of sepak takraw players. To discriminate between high-performance players (HPP), medium performance players (MPP) and low performance players (LPP), the Louvain clustering algorithm was employed. Different SVM models were also developed by varying the hyperparameters of the models. It is evident from the present investigation that anthropometric indexes, particularly standing height, sitting height, leg length, waist circumference, thigh circumference, calf circumference and four-site skinfold measurements evaluated do affect performance in sepak takraw players. It was also demonstrated that the best polynomial-based SVM architecture is capable of discriminating the players with an average classification accuracy of 96% on the validation and test dataset.