Modelos preditores de contratação de profissionais do futebol baseado em Inteligência Artificial por mecanismos Machine Learning

Abstract

This thesis investigates the strategic dynamics involved in the recruitment process of professional football players, with a focus on developing an interpretable decision-support model based on machine learning algorithms, fuzzy logic, and social network analysis. Drawing on an interdisciplinary approach that integrates organisational studies, sports science, and mathematical modelling, the research proposes a methodological solution for evaluating athletes’ technical performance while considering the contextual specificities of contemporary professional football. The study is structured around three analytical axes. The first comprises a bibliometric review, which revealed the emergence of a converging body of literature between sports performance, fuzzy models, and artificial intelligence. The second axis applies social network analysis to over 100,000 international player transfers, identifying patterns of centrality, intermediation, and community formation within the global transfer market—highlighting the strategic roles of countries such as Belgium, Portugal, and Brazil. The third axis presents the development and testing of a fuzzy logic-based performance evaluation model, fuelled by statistics from the Wyscout platform and validated through attributes from Football Manager 2024. The model demonstrated robustness in functional comparisons of players, adhering to the principle of representative design and integrating the tenets of the ecological approach to sports performance. The results indicate the practical usefulness of the tool in supporting decision-making, particularly in talent scouting and efficient resource management. The thesis offers both a theoretical and applied innovation by introducing a flexible scoring system that confronts the subjectivity of traditional scouting methods with analytical criteria, thereby promoting greater strategic rationality in squad composition. Furthermore, the contributions of this work extend to the field of organisational research, by demonstrating how analytical capabilities can become strategic assets in the pursuit of sustainable competitive advantage in global football.