Feature Comparison and Alignment Based Graph Neural Network Resource Recommendation Method

This With the rapid development of online education platforms, the problem of information overload and uneven quality caused by massive learning resources has become increasingly prominent. The traditional recommendation system neglects the interference of semantic information and data noise, which makes it difficult to meet the demand for personalization and accuracy of educational resources. In this paper, we propose a graph neural network resource recommendation method based on feature contrast alignment (RHCLRec), which optimizes the recommendation quality by integrating semantic enhancement and interactive feature alignment through a hybrid contrast learning framework.