Online English Resource Recommendation System Based on Graph Convolutional Network (GCN) Algorithm

The Factor Trans-Convolutional Network (FT-CN) proposed in this paper is an innovative network model. It combines Matrix Factorization (MF), Graph Convolutional Network (GCN) and Transformer to apply to online English resource recommendation system based on GCN algorithm. The model uses MF technology to decompose the interaction matrix between users and English resources into a low-dimensional matrix to obtain the implicit vector representation of users and resources. Then, the graph structure features of English resources are extracted by using the graph convolution operation of GCN combined with the relation between English resources. At the same time, Transformer's self-attention mechanism has been introduced to enhance the model's ability to capture long-distance dependencies and improve the accuracy and efficiency of recommendations. In the experimental part, we compared the FT-CN with the traditional recommendation model. The results show that FT-CN model can make full use of users' behavioral data, content features of English resources and their correlation to provide users with personalized online English resource recommendation services. This fully verifies the feasibility and effectiveness of this framework in the online English resource recommendation system based on GCN algorithm.