Optimizing badminton training plan with artificial intelligence assisted system: A preliminary study

Sports today are heavily reliant on modern technology, making it significant to device efficient methods for receiving valuable information from data. Because machine learning (ML) procedures can handle large datasets, they have been shown to be useful in the treatment of biomechanical data. Sympathetically, these aspects are essential to exploiting training in badminton, where quick reflexes, agility, and accurate actions are significant for presentation. To recover training and assist coaches and athletes in receiving the best out of their training plans, this development aims to develop a system that employs artificial intelligence (AI) to classify among beginner and expert badminton players. Using wearable sensors in a cross-sectional study in a badminton training center, assembly anthropometric and biomechanical data are achieved from both skilled and beginner players. Movement data, such as shuttlecock speed, muscle activity, and footwork dynamics, were recorded by these sensors. Adaptive jelly fish searches Optimized extreme gradient boosting (AJFSO-XGB) method was trained on this data, with the most precise model being further refined through hyper parameter tuning. It is interesting to observe that the AJFSO-XGB technique showed a significantly higher capacity to categorize expert players, an ability that is essential for customizing practices and tactics to enhance performance. These results indicate that improving player growth and optimizing badminton practices can achieve accuracy of 97.05%, F1-scoreof 97.23%, precision of 97.52%, recall of 97.27% and training time of 350 (sec) with the help of an AI-assisted system.