Research on Personalized Learning Path Recommendation System for Adult Education Based on Deep Learning

Aiming at the problems of fragmented course selection and inefficient study planning in computer science majors in colleges and universities, an intelligent learning path recommendation system based on deep learning is designed and developed. The research adopts PyTorch framework to construct a recommendation algorithm model, and realizes personalized learning path generation by analyzing the learning behavior data of 2847 students. The experimental results show that the course completion rate is increased to 89.7% after the system is applied, the exam passing rate reaches 93.5%, and the user satisfaction is 85.3%. The research results provide new technical solutions and practical references for teaching informatization in colleges and universities.