A Novel Adaptive Learning Systems for Development Digital College Chinese Textbooks Based On Bi-directional Gated Neural Networks

The infusion of digital technologies into higher education has significantly reshaped traditional teaching methodologies, especially in language instruction. However, current digital textbooks used in College Chinese often lack the flexibility to adapt to varying student proficiencies and learning preferences. These limitations restrict personalized learning experiences and reduce engagement in today’s digital learning environments. To overcome these challenges, this study proposes an AI-driven framework for the development of adaptive digital textbooks tailored to College Chinese courses. Utilizing advances in deep learning, the system incorporates a Bidirectional Gated Recurrent Unit (Bi-GRU) network to analyze, allowing for real-time content adaptation. The model is trained using the THUCNews Chinese Text Classification Dataset, comprising approximately 840,000 news articles across 14 categories, which serves as a robust linguistic foundation for modeling language comprehension. Experimental outcomes depicts that the proposed Bi-GRU-based system outperforms traditional static digital textbooks, achieving an accuracy of 97.6%, precision of 96.3%, recall of 96.7%, F1-score of 96.5%, and an AUC-ROC of 0.981 in personalizing learning content. These findings highlight the promise of AI-enhanced digital textbooks in delivering more effective, engaging, and student-centered educational experiences in College Chinese instruction.