Using machine learning in Football

Bc. Thomas Stvarnik

Master's thesis

Abstract:

This thesis explores the application of machine learning techniques to football data analysis, focusing on the history, use cases, and implementation of these methods in the sport. The study begins with a literature review, tracing the development of data analysis in football from early beginnings with Charles Reep to the modern age, and discussing the role of event data in the sport. The thesis then delves into the history, types, and workflow of machine learning, with emphasis on supervised, unsupervised, semi-supervised, and reinforcement learning. The practical part of the study is dedicated to the implementation of a machine learning model, specifically the training process and utilization of the final selected model to create a shot map. This implementation relies on Python, XGBoost, Pandas, and Matplotlib technologies. The thesis outlines the intents, workspace configuration, and training data features, before describing the model training algorithm and methods for finding the best results. The study offers valuable insights into the potential of machine learning applications in football data analysis, and enhancing understanding of the game.

Keywords

Machine Learning, Data analysis, Event data, Python, xGboost, Football, xG

Language used: English

Date on which the thesis was submitted / produced: 3. 4. 2023

Thesis defence

• Date of defence: 23. 5. 2023
• Supervisor: Ing. Petr Hanzlík, Ph.D.
• Reader: Ondřej Gojda, Ph.D.

Czech University of Life Sciences Prague

Faculty of Economics and Management

Master programme:
Informatics