Using the CBR dynamic method to correct the generates learning path in the adaptive learning system

Abstract

The adaptive learning systems have the capacity to adapt the learning process to the needs/the rhythms of each learner, the learning styles and the preferences, but they do not ensure an individualized follow-up in real time. In this article, we will present our architecture of an Adaptive Learning System using Dynamic Case-Based Reasoning. This architecture is based on the learning styles of Felder-Silverman and the Bayesian Network to propose the learning path according to the adaptive style and on the other hand on the approach of the Dynamic Case-Based Reasoning to ensure a prediction of the dynamic situation during the learning process, when the learner has difficulty learning. This approach is based on the reuse of past similar experiences of learning (learning path) by analyzing learners’ traces.