IJAIEM

Saurabh Kumar, Pudari Tharun , Duddeda Bharath , Dr. B Laxmi Kantha

Abstract

Injuries in professional soccer are a major problem for both players and clubs. Serious injuries have the potential to cause detrimental effects on a player’s career, or even end it prematurely. Clubs also suffer when key players are injured; their game tactics may need to be changed to better fit the limited team members available, which can severely affect performance, especially in competitive leagues. The goal of this research is to demonstrate the feasibility of using a large dataset to train an accurate Machine Learning model to predict injuries in professional-level soccer. Machine Learning algorithms used to solve this problem have historically had small datasets, which are prone to variance and unreliable results. Therefore, a larger dataset will be constructed from publicly available data to prove that a reliable injury prediction tool can be created. The data in this study will span multiple years and include data about player minutes, age, appearances, and whether the