AI-DRIVEN PREDICTION FOR GOAL SCORING IN FOOTBALL

This paper is dedicated to the development of a machine learning model for predicting the probability that a football attack will result in a goal. The goal is to provide clubs with a tool to strengthen their strategies and increase their chances of better performance, thus improving the financial performance of the team. To achieve this goal, tasks were set: the collection and processing of data, training a machine learning model to predict the probability of a goal. During the study, a hypothesis was put forward that artificial intelligence can accurately predict the outcome of an attack. To test this hypothesis, various algorithms were used, such as logistic regression, random forest, and gradient boosting. The results of the study showed that the machine learning model has a high predictive ability. The results of the study can be used by clubs to improve their strategies. This can lead to an improvement in the financial performance of the team, as winning games and achieving good results in competitions is crucial for financial success, especially for small clubs with limited transfer income.

Keywords:machine learning, predictive modelling, forecasting, data science in sports