balance control ability evaluation & center of pressure (cop) classification with machine learning methods

Athletes' balance control ability is essential in different sports. effective analysis of athletes' balance control ability is an effective way for coaches and sports teams to identify subjects' skills. in the last few years, with the rapid growth of technology in sports, the necessity of using intelligent methods has increased. this study compares different artificial intelligence approaches to evaluate balance control ability by processing time-series data from the center of pressure. a recording pad collects center of pressure data from four types of subjects, ranging from professional skiers to non-athletes. several experimental feature-extraction techniques were applied to the data, and the resulting features were used as input for artificial intelligence methods. this paper utilizes a multi-layer perceptron to classify subjects' skill levels. compared with other methods, the multi-layer perceptron achieves more than 92% accuracy in classifying subjects' proficiency, yielding the best performance. other methods, including k-nearest neighbors and support vector machines, achieved 72% and 69% accuracy, respectively. analysis of center of pressure data can help identify promising individuals for real-world applications.