Thesis Details
Intelligent Manager of Fantasy Premier League Game
Fantasy Premier League online game gives millions of players around the world the chance to become a manager of their club for a while. The results and scores in the game depend on correctly predicting how players will behave in real football matches. If the user had software for predicting and analyzing players' future performance, it would help with making decisions and the outcome of the game could significantly improve. This master's thesis deals with the design and implementation of a prediction model that uses neural networks for time series prediction throughout the game season. Various methods were used to process player and club data for the last 4 seasons. The results are presented in the form of a web application where users can use the created model on their teams. Performance and accuracy of prediction methods were tested on the data from the last season of the Premier League and algorithm predictions were in most of the cases close to reality. If the user used the prediction model's advice 100% in the game, he would score more points than a regular player who does not use any prediction model.
Fantasy Premier League, Premier League, fantasy sports, football, prediction, neural networks, machine learning, supervised learning, time series forecasting, web application
Bartík Vladimír, Ing., Ph.D. (DIFS FIT BUT), člen
Burget Radek, doc. Ing., Ph.D. (DIFS FIT BUT), člen
Češka Milan, doc. RNDr., Ph.D. (DITS FIT BUT), člen
Chudý Peter, doc. Ing., Ph.D. MBA (DCGM FIT BUT), člen
Smrčka Aleš, Ing., Ph.D. (DITS FIT BUT), člen
@mastersthesis{FITMT23053, author = "Maro\v{s} Vasili\v{s}in", type = "Master's thesis", title = "Intelligent Manager of Fantasy Premier League Game", school = "Brno University of Technology, Faculty of Information Technology", year = 2020, location = "Brno, CZ", language = "english", url = "https://www.fit.vut.cz/study/thesis/23053/" }