An Intelligent Classification Method for Online Chinese Language Teaching Resources in Higher Education Based on Deep Reinforcement Learning

In modern higher education, especially in college Chinese Teaching (ChT), online resources play a crucial role in enhancing teaching content and effectiveness. However, managing and utilizing these resources efficiently remains a challenge, particularly in data classification. Traditional classification algorithms often lack accuracy and struggle to handle large, complex datasets. This study addresses these issues by applying Deep Reinforcement Learning (DRL), specifically a Deep Q-Network (DQN), to improve the classification of online ChT resources in universities. To enhance performance, an optimized reward mechanism is designed to automatically learn the best classification strategy for extensive teaching materials. Experimental results show that the proposed DRL-based approach outperforms competitive classification techniques across multiple performance metrics. This research not only introduces an innovative method for managing online ChT resources but also expands the application of DRL in education.