A carregar...
Publicação
Leveraging Machine Learning for Injury Prediction and Prevention in Professional Football: A Machine Learning-Based Framework Using Design Science Research
| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 1.32 MB | Adobe PDF |
Orientador(es)
Resumo(s)
The occurrence of injuries remains a major problem in professional football which affects both
team performance and financial planning and athlete career duration. The research established
a functional injury prediction system through machine learning that combined workloads and
physiological factors with pitch conditions and weather elements and competition levels. The
research used a systematic approach to unite a comprehensive literature review with
experimental work on a well-organized football dataset. The research team performed
systematic preprocessing through feature engineering and encoding and SMOTE application to
handle class imbalance before testing multiple algorithms. XGBoost achieved the optimal balance
between recall and F1-score through GridSearchCV tuning which proved essential for identifying
actual injury risks because missing a case would result in substantial costs. The approach became
applicable in elite sports through SHAP interpretability tools which provided clear explanations
about the factors influencing each prediction to help coaches and medical staff make decisions.
The final model both identified players at risk and provided clear explanations to enable specific
interventions instead of general approaches. The study shows that using internal player metrics
together with external match conditions within a strong machine learning framework produces
improved injury forecasting results. The research creates a strong base for additional
improvements including real-time tracking data integration and team-wide model deployment
to advance individualized injury prevention in professional football.
Descrição
Dissertation presented as the partial requirement for obtaining a Master's degree in Data Science and Advanced Analytics, specialization in Data Science
Palavras-chave
Machine Learning Football Injury Prediction Player Health SDG 3 - Good health and well-being SDG 9 - Industry, innovation and infrastructure