HAL will be down for maintenance from Friday, June 10 at 4pm through Monday, June 13 at 9am. More information
Skip to Main content Skip to Navigation

Maximizing students' engagement through effort-based recommendations

Barbara Moissa 1
1 BIRD - Building artificial Intelligence between trust, Responsibility and Decision
LORIA - AIS - Department of Complex Systems, Artificial Intelligence & Robotics
Abstract : Data exploitation is a growing phenomenon that is present in different scenarios, including the educational scenario, where it holds the promise of advancing our understanding and improving the learning process. From this promise emerged the learning analysis research field that, ideally, takes advantage of technology and educational theories to explore the educational data. On the technological side, we are interested in recommendation systems because they can help students, teachers and other stakeholders to find the best learning resources and thus achieve their learning goals and develop competencies in less time. On the theoretical side, we are interested in the social influence technique foot-in-the-door, which consists in making consecutive requests with an increasing cost. This technique seems particularly relevant to the educational context because it can not only be formalized into a recommendation system, but it is also compatible with the zone of proximal development that states that the challenge presented by the learning resources need to increase gradually in order to keep students motivated. However, we do not know to what extent explicitly applying this technique via recommendations can influence students. Therefore, in this thesis, we investigate such influences assuming that students’ effort is a good indicator of the cost of the requests, since not only every learning activity requires a certain level of effort and, but it is often cited as a key factor for students’ success. For this, we modeled the measurement and prediction of the students’ effort through machine learning models using data that can be used in real life and exploited it in order to explicitly apply the foot-in-the-door technique in a recommendation system. Our results show that, compared to recommendation models that do not formalize this technique, the proposed recommendation models have a positive influence on the students’ effort, compliance, performance and engagement. This suggests that this approach has the potential to improve the learning process as students will present the aforementioned behaviors.
Complete list of metadata

Contributor : Thèses Ul Connect in order to contact the contributor
Submitted on : Thursday, February 10, 2022 - 10:05:41 AM
Last modification on : Saturday, February 12, 2022 - 3:37:53 AM
Long-term archiving on: : Wednesday, May 11, 2022 - 6:18:20 PM


Files produced by the author(s)


  • HAL Id : tel-03564001, version 1


Barbara Moissa. Maximizing students' engagement through effort-based recommendations. Computer Science [cs]. Université de Lorraine, 2021. English. ⟨NNT : 2021LORR0230⟩. ⟨tel-03564001⟩



Record views


Files downloads