This experiment analyzes the ski sensory data, and builds a logistic regression model to classify skill levels: pro and intermediate.
Traditionally, ski trainers visually inspect the movements of skiers in order to evaluate the performance of a skier and determine the areas that the skier needs to improve. Such evaluation is qualitative, and in many cases it is time consuming for both the trainers and the trainees. \n\nIn this study, sensors are attached to 9 different body parts of the skiers and on the skis. Data is collected at a high frequency from both professional and amature skiers when they are completing some tasks. These sensors collect information like position, speed, torque, etc.\n\nThis experiment analyzes and processes the sensory data, and build logistic regression model to classify the skill levels of athletes. It extracts features from the raw sensory data to characterize the movements and coordinations of different body parts. \\nSuch analysis can be helpful to understand the major difference between athletes at different skill levels. It can be very valuable for sports trainers to quantitatively know the areas of improvements, and provide trainings accordingly.