Predicting the transfer value of players is a complex task that requires a combination of football knowledge and forecasting ability. However, with the advancement of technology and the use of machine learning and deep learning methods, predicting the transfer value of football players has become more feasible and promising.
Our project will utilize football player data to construct a model for predicting transfer fees. Each player's data encompasses fundamental information (age, height, playing position, etc.), data about the performances on the pitch (number of goals, assists, injuries, etc.), individual and team awards for the player. The target attribute we will employ is the transfer value of players based on Transfermarkt's estimations.
Transfermarkt is a website specialized in data and information about players, clubs, and the transfer market. Transfermarkt's player valuation system is highly reliable and widely recognized, being extensively consulted by professionals and football enthusiasts worldwide. Subsequently, we will employ machine learning techniques to make predictions about a player's transfer fee and then compare these predictions with the target attribute to assess the model's accuracy.
Effectively predicting the transfer value of players will provide football teams with a more detailed and clear perspective when making transfer-related decisions. The prediction outcomes can be utilized to support the investment decisions of clubs, helping them to recruit valuable players at reasonable prices or avoid acquiring overpriced players with insufficient capabilities.