Predicting physical activity levels from kinematic gait data using machine learning techniques

Svonko Galasso, Renato Baptista, Mario Molinara, Serena Pizzocaro, Rocco Salvatore Calabrò, Alessandro Marco De Nunzio

Abstract: Highlights•Prediction of the physical activity level via kinematic gait data.•Extraction of relevant statistical kinematic features to predict physical activity.•Classification of models’ prediction performance based on optimal features space.•Identification of optimal motion sensors set-up for physical activity prediction.•Classification of the predicting models based on the ranking of kinematic features.