E-learners grouping in uncertain environment using fuzzy ART-Snap-Drift neural network

Abstract

Personalizing the learning contents and programs to each learner is one of the most important goals of e-learning. So, a system should be designed for assigning appropriate learning objects to each learner based on his/her needs abilities and preferences. Automatically grouping the learners in homogeneous groups is an important subject in designing the adaptive learning system. In this paper a new method based on Fuzzy neural network for e-learners grouping is proposed. This new neural network is like to ART network in architecture and Snap-Drift network in learning mechanism. The performance of the network is monitored by a new defined energy-like function. Then, an appropriate learning mechanism is selected in each epoch. Consequently, a high performance network in non-stationary environment is designed. For evaluation of this method, E-Learners of the C programming course are grouped by the proposed method based on Felder-Silverman learning style index. The result of this evaluation shows that our method has appropriate performance in P&G indexes. According to the experimental results, this method has a good performance in uncertain and noisy input environment.