The design of an intelligent multi-agent system for supporting collaborative learning

Students who work in groups in groups encourage each other to ask questions, explain and justify their opinions, articulate their way of thinking, elaborate and reflect upon their knowledge. The benefits of collaborative learning platform are only achieved by active and well-functioning teams. But, can a human tutor support a large quantity of information stemming from a lot of learners interactions? Can the tutor send recommendations and remarks to each student? Is it possible for the tutor to intervene in the due time? Considering these constraints in terms of feasibility, tutor availability, and time latency, the purpose our project is to automate the human tutor tasks. In this paper we present the design of an intelligent multi-agents system that assists learners during collaborative E-learning. The assistance is done through the analysis of the cognitive and social indicators which determine the outcomes of interactions between learners to estimate their collaboration state. Based on these indicators, the system generates automatically some recommendations that are suitable for every learner. The automation of the tutor roles is achieved by an agent that uses fuzzy logic techniques for its machine learning.

1. Introduction

Collaborative learning is any learning activity that is realized by a group of learners who have a common purpose, each student should be source of information, motivation, interaction, mutual aid. Also, it is an activity where learners benefit from the contributions of others and the help of a tutor whose job is to facilitate the individual and collective learning [1]. The problem that arises in the case of E-learning, it is how to support the collaborative work taking into consideration the separation of space-time between learners themselves on one hand, and between learners and tutor from the other hand.