Technology-Enhanced Learning of Motions Based on a Clustering Approach

The analysis of user motions can be useful in many fields to observe human behavior; to follow and predict its action, intention, and emotion; to interact with computer systems; and to enhance user experience in virtual (VR) and augmented reality (AR). These analyses can be empirically made by the expert or with the help of a technology-enhanced learning (TEL) system, allowing the extraction of relevant information from the motion in a pedagogical context. Such analyses are rarely made from 3D captured motions. This can be explained by several factors: the complexity and high dimensionality of the data and the difficulty to correlate the observation and analysis needs of the expert to the extracted data. Machine learning techniques could be used to address some of these problems. In particular, the use of unsupervised learning techniques could help in giving advice according to the analysis of clusters, representing user profiles. During a learning situation, the expert will be assisted in their evaluation task. This work presents two main contributions: (i) the use of clustering techniques to separate motions, into different categories according to a set of well-chosen features, and (ii) the development of a TEL environment using clustering techniques in order to assist the expert in its motion-based evaluation task.