Publicação
FPF field lab: predicting the attendance of portuguese 1st division football stadiums using machine learning
| Resumo: | This paper aims to help FPF predict attendance at Portuguese Liga I according to data from the 2015/16 season until 2021/22 to optimize round scheduling of football matches. The dataset contains information about games played, including the teams, stadiums, and weather conditions. Among all the various machine learning regression models tested, the XGBoost provided the overall best results regarding the out-of-sample mean absolute error. Additionally, considering the differences in size and consequently in error between two groups of clubs, narrowing down the scope added value to the project. Besides attendance, the occupation rate was also tested as the target variable. |
|---|---|
| Autores principais: | Passeiro, Diogo |
| Assunto: | Machine learning Prediction Stadium attendance Football |
| Ano: | 2022 |
| País: | Portugal |
| Tipo de documento: | dissertação de mestrado |
| Tipo de acesso: | acesso aberto |
| Instituição associada: | Universidade Nova de Lisboa |
| Idioma: | inglês |
| Origem: | Repositório Institucional da UNL |
| Resumo: | This paper aims to help FPF predict attendance at Portuguese Liga I according to data from the 2015/16 season until 2021/22 to optimize round scheduling of football matches. The dataset contains information about games played, including the teams, stadiums, and weather conditions. Among all the various machine learning regression models tested, the XGBoost provided the overall best results regarding the out-of-sample mean absolute error. Additionally, considering the differences in size and consequently in error between two groups of clubs, narrowing down the scope added value to the project. Besides attendance, the occupation rate was also tested as the target variable. |
|---|