Based on Bayes theorem, naive Bayes classifier has stable performance. For different data sets, the classification effect is not different, and the model has good robustness. Based on naive Bayes classifier, this paper studies the algorithm and system of learning resource recommendation. Firstly, based on the method of use case diagram, the use case diagram of student user, teacher user and system administrator user are constructed. Then, we study the algorithm of naive Bayes classifier, including Gaussian naive Bayes model, polynomial naive Bayes model and Bernoulli naive Bayes model. Finally, three kinds of analysis algorithms are simulated, and the experimental results show that the Gaussian naive Bayes model has advantages. This system uses the Gaussian naive Bayes model to build the classifier and realize the recommendation of learning resources.