The design of innovation and entrepreneurship learning platform based on graph neural network recommendation algorithm

The emergence of new technologies such as artificial intelligence, big data, and cloud computing provides richer resources and means for the development of innovation and entrepreneurship education, driving innovation and entrepreneurship learning platforms to adapt to rapidly changing market demands. The study proposes using graph neural network recommendation algorithm to construct a platform system, representing users' knowledge level and ability through a group model. Secondly, a learning generation network is constructed to capture feature vectors to design a hybrid recommendation algorithm. The results indicate that the maximum accuracy of this recommendation algorithm exceeds 0.92, which is significantly better than other comparison algorithms. And the maximum delay value of this recommendation algorithm on the system does not exceed 1.5ms, and the maximum information query rate is stable at 1320K req/s. The application load is better than other comparison algorithms. The designed learning platform system can effectively achieve information push and functional query, meet users' personalized needs, and improve their practical abilities.