Improving University Course Recommendation Via Knowledge Graph and Attention Mechanism Enhanced Gated Recurrent Unit

In recent years, the quality of education has improved rapidly, and the types of university courses have also increased. However, in the face of a wide variety of courses, students do not know how to choose, and some high-quality and unpopular courses are not chosen. In view of this, this paper proposes a university course recommendation model that integrates knowledge graph and attention mechanism to improve gated cycle unit. The experimental results show that when the number of iterations is 800, the prediction accuracy of this model is $97.1 \%$, which is better than $93.8 \%, 89.3 \%$ and $87.1 \%$ of other models. The hit rate and recommendation quality evaluation values are both maintained at about 0.9, which is significantly better than about 0.5 of other models. The experimental results show that the model has excellent performance in accuracy rate, hit rate and recommendation quality evaluation. To sum up, this study puts forward a recommendation model for college students’ course selection, which can meet students’ interest needs, enhance students’ comprehensive strength, and give play to the important value of college courses.