Corresponding author: Rochdi Boudjehem ( boudjehem.rochdi@univ-guelma.dz ) Corresponding author: Yacine Lafifi ( lafifi.yacine@univ-guelma.dz ) © Rochdi Boudjehem, Yacine Lafifi. This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY-ND 4.0). This license allows reusers to copy and distribute the material in any medium or format in unadapted form only, and only so long as attribution is given to the creator. The license allows for commercial use. Citation:
Boudjehem R, Lafifi Y (2021) A new approach to identify dropout learners based on their performance-based behavior. JUCS - Journal of Universal Computer Science 27(10): 1001-1025. https://doi.org/10.3897/jucs.74280 |
Distance learning environments are increasingly offering more comfort to both learners and teachers, allowing them to carry out their academic tasks remotely, especially in critical times where it is difficult, or even dangerous, to bring these actors together in one physical place. Nevertheless, These same environments are complaining about the massive dropout numbers among their learners. Therefore, designing new intelligent systems capable of reducing these numbers becomes imperative. This paper proposes a new approach capable of identifying and assisting endangered learners experiencing difficulties by monitoring and analyzing their behavior inside the e-learning environment. By building dynamic models to follow the learners’ current situation, the proposed approach could intervene autonomously to save learners identified as struggling. Relying on distributed artificial intelligence instead of humans to closely monitor learners within distance learning environments can be very effective when identifying struggling learners. Furthermore, targeting these learners with early enough and carefully designed interventions can reduce the number of dropouts.