Classification of Foot Kicks in Taekwondo Using SVM (Support Vector Machine) and KNN (K-Nearest Neighbors) Algorithms

Abstract

Taekwondo is one of the most popular martial arts in Indonesia. The number of competitions held at various regional levels proves that taekwondo is popular. In an online match, judges assess movement through video so that errors can occur in judging the type of kick. No research discusses the classification of kick types. However, several related studies examine monitoring human movement using SVM and KNN algorithms. This experiment aims to determine the performance of the SVM and KNN algorithms in classifying taekwondo foot kicks. The experiment results show that the SVM and KNN algorithms effectively classify taekwondo foot kick types. Using the kurtosis feature, the SVM algorithm obtained an accuracy rate 91.6%. As for KNN, the accuracy rate is 96.8% with the kurtosis feature.