PREDIKSI MARKET VALUE PEMAIN SEPAK BOLA DI LIMA LIGA TOP EROPA MENGGUNAKAN K-NEAREST NEIGHBOR

Abstract

Every football club in competition at any country has ambition to be a champion. One of the important things to be a champion is the quality of player in the team. Therefore, when transfer time is opened, almost every club buys the required players. Busy transfer activity frequently occurs in five Europe top leagues. Nevertheless, a club frequently buys a player which has price beyond market value transfer (market value). The reason is that many clubs still don not know how to determine market value. The aim of this research is to predict the market value of football players in five Europe top leagues using K-nearest neighbor (KNN). Data which will be used are 26 football players who play in five Europe top leagues with club get into first tier league (promotion) in season of 2022/2023. The result is that the model which is used for prediction is a KNN model which has proportion of training data and test data at the ratio of 90:10 and parameter K = 3 since it has the lowest MAPE which is 10,45 %. From model selection we obtain MAPE value at 28,614 % for compared prediction of market value result with actual of market value.