Lógica fuzzy e o Poker : a Matemática para vencer a sorte

• Ano de defesa: 2024
• Autor(a) principal: Branquinho, Ana Clara Boscolo
• Orientador(a): Silveira, Graciele Paraguaia
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de São Carlos; Câmpus Sorocaba
• Programa de Pós-Graduação: Programa de Pós-Graduação em Ensino de Ciências Exatas - PPGECE
• Palavras-chave em Português: Poker; Classificação de Perfis; Modelagem Matemática; Lógica Fuzzy
• Palavras-chave em Inglês: Profile Classification; Mathematical Modeling; Fuzzy Logic
• Área do conhecimento CNPq: CIENCIAS EXATAS E DA TERRA::MATEMATICA
• Link de acesso: https://repositorio.ufscar.br/handle/20.500.14289/20500

Poker is a highly popular card game worldwide, involving aspects such as strategy, technical skill, risk management, and luck, and is currently considered a mind sport. Due to the existence of monetary prizes for winners, the inherent complexities of this game are extensively studied to develop strategies aimed at improving the performance of players in high-level competitions, many of which are grounded in logic and mathematics. The objective of this dissertation was to develop a computational mathematical model based on fuzzy logic that classifies poker player profiles, specifically for the Texas Hold’em Online variant. The proposed methodology involves analyzing statistical data from players and consists of fuzzy modeling of critical variables related to the dynamics of the game. Inferences and defuzzification are performed using Mamdani and Center of Gravity methods and are computationally processed with the MATLAB fuzzy toolbox. The player data was extracted from the Poker Tracker platform, and access was facilitated through association with a professional player. Additionally, research and tests were conducted to determine standard reference values, ranges for class definitions, and validation of the results returned by the developed model. It is concluded that the model can predict actions a player might take at certain moments in the game and also their chances of success. This approach deepens the understanding of the complex