A NOVEL APPROACH FOR PREDICTING FOOTBALL MATCH RESULTS: AN EVALUATION OF CLASSIFICATION ALGORITHMS

Abstract

The task of predicting the outcomes of football matches is rendered increasingly complex by the intricate nature of the game and the variety of variables that could affect how things turn out. In the recent past, machine learning algorithms have been applied to this challenge, with varying degrees of success. In this particular research paper, we have meticulously evaluated the performance of several classification algorithms with the objective of predicting the outcomes of football matches in a tournament setting. The algorithms that were thoroughly tested encompassed a diverse range of classification models, including logistic regression, support vector machines and random forests. The study employed a dataset of historical match data drawn from the FIFA World Cup, historical team ranking data and team strength data from FIFA games. In order to accurately assess the efficacy of the algorithms tested, the evaluation metrics used were accuracy, precision and recall. The results of the study highlight the fact that machine learning algorithms can indeed prove to be effective tools for predicting the outcomes of football matches.