Improving INGInious : labeling mechanism to better identify difficulties of students

Details

Supervisors

Bonaventure, Olivier

Faculty

Ecole polytechnique de Louvain

Degree label

Master [120] en sciences informatiques

Abstract

This master thesis covers the implementation and the utilization of a labeling mechanism in the INGInious platform in order to better find difficulties of students. Technologies evolve every day in many domains and education is one of these domains. Courses begin to have online contents such as interactive syllabuses or exercises in the cloud that can create a lot of interesting data to process in order to improve courses and detect difficulties of students. However, research shows that "there is a lack of tools combining both orchestration and evaluation in order to detect missing skills" and that many difficulties encountered by students come from the lack of their prior knowledge. Being able to detect difficulties of students is essential in order to help them in a more targeted way and to create courses with an appropriate level of difficulty. INGInious is an online automatic code assessment platform currently used at Université catholique de Louvain. This work aims to improve this platform to allow teachers to easily mark and record what students are doing and mistakes they make when doing exercises on this platform and then, compute statistics based on this. The implementation has already been used to reveal difficulties of students in several courses given at Université catholique de Louvain.