Research on Accurate Recommendation of Learning Resources based on Graph Neural Networks and Convolutional Algorithms

Abstract

In response to the challenges of "learning confusion" and "information overload" in online learning, a personalized learning resource recommendation algorithm based on graph neural networks and convolution is proposed to address the cold start and data scarcity issues of existing traditional recommendation algorithms. Analyze the characteristics of the Knowledge graph of learners and curriculum resources in depth, use the graph Auto encoder to extract the auxiliary information and features in the Knowledge graph and establish the corresponding feature matrix, and use Convolutional neural network for classification and prediction. The experimental results show that this algorithm improves the performance of recommendation systems, improves learners' learning efficiency, and promotes personalized development.