Evaluación y aplicación de algoritmos de inteligencia artificial explicada para predecir el desempeño de los clubes de fútbol basado en las características individuales de sus jugadores.

Abstract

The sports sector has been one of the main sectors in which data analysis and the application of machine learning models have been implemented to evaluate the performance of several of its areas of interest in recent years. This project analyzes the application of Machine Learning (ML) algorithms and Explainable Artificial Intelligence (XAI) techniques in the context of soccer match score classification based on player characteristics of Premier League teams. The text discusses the selection of algorithms based on their ability to be analyzed by XAI tools such as SHAP and LIME, and their suitability to the type of data available. The test plan for the evaluation of ML algorithms considers the use of confusion matrices. and metrics such as Accuracy, Recall and F1-Score for a more thorough evaluation. Finally, instances of soccer match results were analyzed using XAI techniques and explanatory graphs were obtained detailing which attributes of teams and players affect the winning or non-winning of a professional soccer game.