Optimalizace a simulace kurzových sázek

The topic of the bachelor's thesis is "Optimization and simulation of odds betting". The aim of the work is to create neural network models based on a dataset of tennis matches, which would enable the prediction of match results. The work consists of a theoretical and a practical part. In the theoretical part, key terms from the field of betting, game theory, database systems and neural networks are defined and described. The practical part includes the implementation and creation of the infrastructure of the development environment. First, the data used in the practical part are described here, and then the data are modified for uploading to a database. Next, the practical part explains the procedures for creating and training two models $-$ a static model created using the TensorFlow tool and a dynamic model created using the NEAT neuroevolutionary algorithm. The results of both models are eventually compared to determine their effectiveness and accuracy in predicting the winners of tennis matches. The models are simulated on real betting opportunities to determine the potential profitability of the strategies based on the predictions of these models. This work contributes to the understanding and the use of technological tools for the analysis and optimization of odds betting strategies and can serve as a starting point for future research into the application of machine-learning models.