Techniques and algorithms to predict the outcome of soccer matches using data mining, a review of the literature

Araujo-Ahon A.; Cardenas-Mayta B.; Iparraguirre-Villanueva O.; Zapata-Paulini J.; Cabanillas-Carbonell M. · RISTI - Revista Iberica de Sistemas e Tecnologias de Informacao (2023)

Resumen

The outcome of a sport has become a necessity for competitors as well as for fans following their favorite teams. However, the prediction of the outcome of a soccer match (PSMR) is very varied due to the various existing models. The research is a systematic literature review (SLR) based on manuscripts published in IEEE Xplore, Scopus, Science Direct, and Springer. Prisma methodology was used for analysis and systematization. The objective of this research is to provide a guide for using machine learning (ML) techniques. The results showed that the most frequently used ML techniques are supervised learning (SL) and unsupervised learning (UL) and the most frequent ML algorithm for predicting the outcome of a soccer match is Random Forest (RF), considering its great contribution in prediction accuracy. In addition, a novel and efficient model for predicting the outcome of soccer matches, supported with Data Mining (DM) and focused on ML, is proposed after the study.

Instituciones: Universidad Continental; Universidad Autónoma del Perú; Universidad Privada del Norte