International Journal of Engineering and Technology

This paper proposes a mathematical framework – the SaGe framework, which maps the adaptive nature of the e-learning environment to the reinforcement learning approach. An adaptive e-learning system works in a similar fashion to an intelligent agent that assesses the various interactions of the user with the system, learns from these interactions and uses this knowledge base to select the best possible action to earn the maximum reward. This is the same methodology of reinforcement learning. The adaptive nature of the e-learning environment is provided by the assessment of the individual differences in the learning styles of the students. In our approach we chose the learning styles provided by the dimensions of the Felder-Silverman Learning Style Model (FSLSM) and the SARSA algorithm for the reinforcement learning. Mapping of the reinforcement SARSA algorithm to an adaptive e-learning system is asserted by the time-based update of the Q-values of the SARSA algorithm.