Cooperative learning in self-organizing e-learner communities based on a multi-agents mechanism

Web based teaming provides an unprecedented flexibility and convenience to both learners and instructors. However, it also creates a great number of lonely learners. Hence, there is an urgent need for finding an efficient way to help the learners share their teaming experiences and exchange their learning materials during the teaming process. This paper reports on an attempt to construct a flexible and effective self-organization system that groups similar learners according to their preferences and learning behaviors. We use a multi-agent mechanism to manage and organize learners and learner groups. Furthermore, we present effective award and exchange algorithms, so eventually learners with similar preferences or interests can be clustered into the same community. Experiments based on real learner data have shown that this mechanism can organize learners properly, and has sustainably improved speed and efficiency of searches for peer students owning relevant knowledge resources.