Authors : Hoo-Young Lee, Koo-Rack Park and Joon-Yong Kim
Abstract: Recently, the use of various statistical techniques including IT technology and mathematical technique in establishing strategies and tactics for improving the athletic performance in the sports field has been made around popular events such as baseball and soccer. However, such studies are also needed for unpopular events for the development of society-wide living sports. This study used the Markov chain, a mathematical technique that sequentially records the attack data of a specific player through video reading of the kendo match and then, predicts the future changes based on the status of the past to probabilistically predict the next attack area of the player. This study analyzed the attack data of the kendo match and implemented the predictive system using a Markov chain model and as a result, we obtained the prediction success rate of 78%. It is expected that it will be able to contribute to efficient strategy establishment and match operation of athletes and coaches during the Kendo match. In the future, in order to solve the problem that the player’s intuition is involved according to the defensive posture of the opponent and the flow of the match and a relatively low probability of attack is made, a study to supplement the predictive system by substituting the defense pattern of the defensive player with the prediction formula using the attack data of the attacker should be continued.
Hoo-Young Lee, Koo-Rack Park and Joon-Yong Kim, 2018. Implementation of Kendo Attack Predictive System Using Markov Chain. Journal of Engineering and Applied Sciences, 13: 3860-3864.