Collaborative graph convolution and learning style-based course recommendation method

The invention discloses a course recommendation method based on a graph convolutional network and a learning style, and the method comprises the following steps: 1, predicting a score, coding course embedded information connected with a learner into first-order embedded information of the learner, and obtaining high-order embedded information through the first-order embedded information; aggregating each order of embedded information in the high-order embedded information into a single vector through an aggregation function to obtain an aggregation embedded representation; performing inner product operation on the aggregation embedded representation of the learner and the course to obtain a predicted score of the learner on the course; 2, respectively defining a learner summary and a course summary, calculating a learning style vector, and obtaining a course learning style similarity score according to the learner learning style vector and the course learning style vector; and step 3, optimizing the predictio