Graph convolutional network driven multi-task learning resource recommendation method

The invention discloses a multi-task learning resource recommendation method driven by a graph convolutional network, relates to the technical field of intelligent education, and aims to solve the problem of multi-task recommendation, improve the personalized learning experience of students, optimize reasonable distribution of learning resources and improve the learning efficiency. The method at least comprises the following steps: respectively constructing graph structures of interaction relationships between students and courses, between knowledge concepts and between students and videos; a graph embedding module is used for processing and learning the graph structure by using an improved graph convolutional network and generating a graph embedding representation; taking the graph embedded representation as the input of a multi-task recommendation module, performing feature extraction by using a plurality of expert networks, dynamically integrating expert knowledge by using a gating mechanism, aggregating e