Adaptive deep reinforcement learning for personalized learning pathways: A multimodal data-driven approach with real-time feedback optimization
Shoujiang Ruan, Kebin Lu
Abstract
This article proposes an adaptive online learning platform based on deep reinforcement learning (A-DRL) for intelligent recommendation of personalized learning paths. The aim is to improve learners' learning effectiveness, user experience, and overall engagement. The platform integrates user interaction, learning outcomes, and multimodal data, dynamically adjusting learning paths through deep reinforcement learning technology to automatically adapt to learners' needs and progress. A key feature is the use of adaptive learning strategies to optimize recommendations based on learners' feedback, learning progress, and personalized requirements, facilitating bespoke learning plans. Experimental results demonstrate that the A-DRL-based recommendation system significantly enhances learning effectiveness, user satisfaction, and reduces learning burden. The platform can track learners' behavior in real-time, analyze their emotional and cognitive states, and further optimize learning path recommendations. This study offers an innovative intelligent recommendation framework for online learning platforms and new ideas for personalized education and intelligent teaching systems.