Design of Intelligent Online Education Resource Optimization and Scheduling Strategies Based on Deep Reinforcement Learning

Abstract The increase of intelligent online education resources has brought pressure on learners to choose educational resources, which has become a key issue in the optimization of online education resources. Based on the research of deep reinforcement learning, this paper models the resource recommendation process as a Markov decision model, optimizes the policy neural network, and constructs an online education resource recommendation model based on deep reinforcement learning to improve the recommendation effect of intelligent online education resources. Applying this paper’s model to the recommendation of educational resources in digital libraries, the results of this paper’s model on HR@5, HR@10, NDCG@5, NDCG@10 metrics in the school digital library dataset are 0.83, 0.9222, 0.5901, 0.6219, respectively, which are better than other comparative models. The optimal metrics results are also obtained on the Goodbooks-10k dataset with 0.4807 and 0.7023 for HR@5 and HR@10 metrics, respectively, and 0.3689 and 0.4389 for NDCG@5 and NDCG@10 metrics. To explore the recommendation performance performance of this paper’s model with different parameter settings, this paper’s model, when the number of categories reaches 2000, the HR@5, HR@10, NDCG@5, NDCG@10 indicator results reach the optimum.