Multi-Agent System Based on Fuzzy Logic for E-Learning Collaborative System

Collaborative learning is an active process by which the learner works to build his knowledge. The tutor plays there the role of facilitator of learning while the group participates as a source of information, as a motivational agent. A rapid bibliographic synthesis shows that the evaluation of the quality of collaboration in a group is based on the analysis of the interaction traces left by the stakeholders on the LMS. The analysis of these traces allows tutors to intervene to support and motivate the learners and thus minimize the risk of abandonment. However, because of the significant volumes of traces left by the learners, we found it urgent to develop an automatic system of assistance for tutors concerning monitoring learners to improve their level of involvement in the collaborative process. The system (SISAC) that we have proposed in the context of this research is an intelligent MAS based on the fuzzy logic technique for assessing the level of learner collaboration. Fuzzy system inputs are indicators derived from the analysis of learner traces on online collaborative learning platforms (LMS). For the validation of the system, we conducted experiments with the learners of the ENSA in Ibn Tofail University, Kenitra. These tests allowed us to measure the positive impact of the system on improving the collaborative behavior of learners.