Web-Based Education Digital System Using Hybrid Deep Reinforcement Learning with Attention Mechanism

The accelerated development of digital technologies has revolutionized the delivery of education, with web-based education systems providing flexible, accessible, and scalable learning opportunities. Present platforms, though, tend to be limited in offering personalized learning experiences because they have static content adaptation, poor tracking of learner engagement, and constrained responsiveness to individual progress. Traditional suggested systems, like rulebased adaptation or simple recommendation engines, cannot dynamically adapt content in real-time, leading to low learning effectiveness and motivation. To deal with the limitations of these challenges, this study suggests a Hybrid Deep Reinforcement Learning with Attention Mechanism (HDRL-A) model for designing and implementing a Web-based education digital system. The HDRL-A solution supports dynamic personalization by continuously examining learner interactions, discovering principal engagement features, and adapting content delivery strategies based on a reward-based feedback system. Attention mechanisms support the model's capacity to concentrate on the most salient learner data to predict with higher accuracy. Experimental testing shows that the suggested HDRL-A model improves on common adaptive learning approaches in terms of engagement rate, learning retention, and completion percentage. Comparative analysis verifies its capacity to provide a smart, responsive, and people-oriented learning experience to open the doors for more efficient, individualized, and data-oriented online learning spaces.