Deep learning in sports skill learning: a case study and performance evaluation

Abstract

Deep learning in sports uses neural networks to evaluate data from sensors and cameras, providing coaches and players insights to enhance training methods and performance. Sports skill development include issues with data availability, trouble interpreting methods for coaching purposes, possible financial constraints for players and regional sports teams. To overcome this, we proposed an Artificial Hummingbird Optimized XGBoost (AHO-XGB) to provide accurate predictions and analysis of an athlete's performance.In this study, the research consists of 20 faculty members and 250 learners from 3 universities.Many sports talents are currently taught to students in famous colleges and universities, but they truly become proficient in the skills. To evaluate the performance of the proposed method in terms of accuracy (92.6%), precision (90.5%), and recall (94.3%). The outcome of this research in sports skill learning transforms performance and training analysis by examining large amounts of data and offering suggestions for skill development.