Analysis of Y-Balance Test Data Used in Sports Sciences with Machine Learning Methods

Abstract

The Y-Balance Test (YBT) is a popular method used to evaluate the dynamic balance and functional mobility of athletes. However, it is crucial to perform the test correctly to obtain accurate results. For precise measurement of YBT data, a separate measurement should be conducted for each individual, and the test grid and equipment must be set up correctly. Clear instructions and a standardized protocol must be followed to avoid misleading results, which can lead to incorrect information being used to design training programs to improve athletes' performance and reduce injury risks. To address this challenge, a study was conducted using supervised learning methods to explore YBT data, perform preprocessing steps, and conduct a comparative analysis of different machine learning models' performance in predicting YBT data. The study predicted YBT values, which require individual measurements, using different machine learning methods based on determining features such as age, gender, and training age. The experimental results demonstrated that the predicted YBT values can aid in designing training programs that can enhance athletes' performance and reduce injury risks. Overall, the findings of the study highlight the importance of accurate YBT data measurement and the potential of machine learning methods in predicting YBT values based on an individual of specific features. This approach can provide valuable insights to coaches, trainers, and healthcare professionals to create tailored training programs that can improve athletes of balance and mobility while minimizing injury risks.