Service design with machine learning based on user action history(Comparison and Visualization of differences in running motion with Dynamic Time Warping)

With the development of IoT techniques and rising attention on individual service, consumers and producers will mutually exchange their intelligence and better customize product development processes. This study focuses on the user's daily life, examines a proposed system using sensor shoes with several sensor devices embedded in the insoles, collecting action data of users, extracting their action features, and then issuing some advice based on the difference of action features in different situations to help users train more efficiently. In the previous works, a service model uses a Recurrent Neural network (RNN) to distinguish users' actions and to extract their action features using Self-organization Map (SOM) from the presented sensing data and the feasibility of these methods are confirmed. In this paper, a more quantitative and intuitive comparison method, named Dynamic Time Warping (DTW), for the action feature analysis part is proposed. The performance of the proposed method is reviewed via experimentation.