An Adaptive Personalized E-Learning Platform With Cognitive Varwk Adaptation And Gamification

Abstract

In recent decade, effective and engaging e-learning solutions are imperative, particularly for the early stages of K-12 education. This paper introduces an Artificial Intelligence (AI) approach to e-learning, presenting an adaptive personalized platform that combines Visual/Aural/Read/Write/ Kinesthetic (VARWK) presentation, exercises difficulty scaffolding through skipping/hiding/ reattempting and gamification elements. The platform employs cognitive, behavioral, and affective adaptation techniques to develop a dynamic learner model, effectively identifying and correcting learning style and cognitive level. The adaptation targets encompass adaptive content presentation (VARWK and gamification), exercises navigation, and feedback. To realize its objectives, the platform utilizes Deep Residual Network Learning (DRNL) and an online rule-based decision-making framework. The platform features a front-end dedicated website interfacing with back-end adaptation algorithms. A experiment are conducted on various grades with English curriculum demonstrates improved learning effectiveness, as evidenced by a comparison between post-test and pre-test results. Both groups of students experienced enhancements in academic performance and satisfaction levels, with the VARK group showing slightly greater improvement and higher satisfaction due to the engaging interactive activities and games in the kinesthetic presentation, while maintaining accessibility to other presentation styles when needed.