This bachelor's thesis examines the use of data mining techniques on betting odds of football matches in the Czech First League spanning from the 2016/2017 season to the 2022/2023 season. Given the very extensive data from previous years, we can use data mining techniques to predict the best possible bets with the goal of profit. The theoretical part focuses on the CRISP-DM methodology, which includes the phrases of problem understanding, data understanding, data preparation, modeling, eval... show full abstractThis bachelor's thesis examines the use of data mining techniques on betting odds of football matches in the Czech First League spanning from the 2016/2017 season to the 2022/2023 season. Given the very extensive data from previous years, we can use data mining techniques to predict the best possible bets with the goal of profit. The theoretical part focuses on the CRISP-DM methodology, which includes the phrases of problem understanding, data understanding, data preparation, modeling, evaluation, and deployment of results. The practical part is structured by this methodology. The introduction of LISp-Miner system is also included in the theoretical part. In the practical part of the work, data mining steps are used on selected data from data acquisition to interpretation of analysis results. Modeling is implemented in LISp-Miner environment, using 4ft-Miner, CF-Miner and SD4ft-Miner procedures, which are used to solve predefined analytical questions. The results of the analysis are then displayed and further analyzed in the MS Office Excel system. The main goal is to find long-term trends in betting odds and to find ideal bets for generating profit. |