PhD fellowship in digital learning support

The purpose of the project is to develop and evaluate the use of digital support in learning. We look at how technology can be used in an intelligent way, by dynamically adapting and personalizing the learning environment to the learner, based on a permanent monitoring of various skills and other person features. To this end, advanced techniques from statistics, psychometrics and computer sciences will be adapted and applied. The project focuses on two application contexts: an industrial training setting and a higher education setting. In the industrial setting, we will focus on an adaptive support system for an effective and efficient use of Augmented and Virtual Reality (AR/VR) in training industrial operators. The project starts by building up a content model that collects information about levels of support and various features of tasks. The system will then collect training data and conduct an operational analysis to make each level of support scalable. Based on the operator’s performance on the tasks, the learner model measures and updates information about the operator’s latent traits (specifically, proficiency, well-being, and learning). In particular, our learner model combined with Elo rating algorithm (Elo, 1978) allows to keep track of the operator’s states. Finally, based on this information on the operator’s states, the recommender engine chooses tailored and timely tasks, and offers personalized support for the tasks assigned to the operator. In the higher education setting, we will set up a dynamic system that supports students to select relevant courses and to track the progress students make. To implement such an assistant system, the PhD candidate will propose an advanced multidimensional domain model that allows to represent the skills (including soft skills) worked on during the education program, the dependence between them and their constraints. A dynamic evolution model, based on the domain model and student’s learning traces, marks and history of courses, is needed to track and predict the evolution of the student’s skills profile. Also, the candidate will propose a system that will assist the student in choosing the sequence of courses to follow in order to reach a final graduation skills profile set by the student himself. The proposed sequence can be adapted as the student evolves during the terms. A failure in the system planning to reach an objective profile should be treated with human assistance to review the orientations of the students’ objectives. Data will not only be used to give recommendations to students, but also to feed a permanent re-engineering process of the courses’ adjustment, dependencies and potentially their content.