Research on Personalized Recommendation Strategy for Teaching Content of Sports Culture Based on Deep Learning

This paper mainly establishes a recommendation model based on deep neural network to realize the personalized recommendation of physical culture teaching content. Through the feature selection method based on MIFS, the learners’ preference for physical culture teaching content features is determined, and the input process of the recommendation method is completed. Then a two-part graph association model is established to visually describe the association relationship between learners and resources. Finally, deep network learning is used to optimize the entire personalized recommendation process. Recommendation performance test can be found, when the number of content N is 5, the deep neural network check accuracy rate is about 40%, compared with other algorithms, deep neural network has better performance. The application of this paper’s personalized recommendation platform for sports culture teaching content can significantly improve sports knowledge, sports awareness and sports behavior (P<0.01), and also have a greater improvement on the performance of college students’ physical fitness level.