Application of different Model Algorithm in the Prediction of Transfer Fee of Soccer Players

Abstract

Soccer is popular worldwide, and the fees for transferring soccer players show how much a player is worth and give us an idea of how well a country’s soccer is developing and how a team is being managed. This study aims to investigate the application of different model algorithms in the prediction of the transfer fee of soccer players and find which model is the most accurate. The dataset used for this research is available on the Kaggle website, from the football player’s transfer fee prediction dataset. By analyzing the (players’ codes and names) football team, position, height, age, the appearance of a player, the number of goals, assists, yellow cards, second yellow cards, red cards, goals conceded, clean sheets, minutes played, days-injured, games-injured, award, current-value, highest-value and position-encoded. Machine learning is commonly used in diverse fields to solve difficult problems that cannot be readily solved in based on computer approaches [1]. This study compares the accuracy of different machine learning algorithms used for predictive analysis of soccer players’ transfer fees. For example: Linear regression, Random Forest, Decision tree, K-Neighbors, and Neural network. When finding the relationship between those 21 factors, could help players and teams to make valuable decisions and accurate prediction for the establishment of soccer player market.