Research on University Textbook Recommendation Method Based on LightGCN

With the rapid development of information technology, artificial intelligence technology has been applied to the selection and recommendation of university textbooks to some extent, but it still faces the problem of low recommendation accuracy. To solve this problem, this paper proposes a university textbook recommendation method based on the LightGCN (Light Graph Convolutional Network) model. This method utilizes the characteristics of graph neural networks, constructs a graph structure between textbooks and teachers, and deeply explores the relationship between teacher preferences and textbook features. First, we construct a user-textbook graph by combining the historical interaction data of teachers selecting textbooks. Then, we use LightGCN for node representation learning to improve the accuracy and efficiency of recommendations. Experimental results show that the proposed method significantly outperforms traditional recommendation algorithms in multiple recommendation indicators, and has good practical value and prospects for promotion. This study provides new ideas and methodological support for the optimization of university textbook recommendation systems.