Personalized Book Intelligent Recommendation System Design for University Libraries Based on IBCF Algorithm

With the digital transformation and improvement of university library information technology, readers' demands for library services are increasingly diversified and personalized. They are no longer satisfied with the traditional borrowing services, but hope that the library can provide more accurate and personalized recommendation services. To solve these problems, this study first proposes an improved item-based collaborative filtering recommendation algorithm based on the mean model representation. Then, combining this algorithm with user-based collaborative filtering recommendation algorithm, an improved item-based collaborative filtering algorithm is designed. The results showed that the CPU usage of the whole system was not high during the operation of the improved item-based collaborative filtering recommendation algorithm, with an average usage rate of about 9.8%. The minimum root mean square error of the algorithm was 0.013 and the runtime was 12000 mu s. Compared with existing the similar systems, when the number of users exceeded 200, the response speed was significantly reduced by more than 50%, and the coverage rate reached more than 90%. In summary, the personalized intelligent book recommendation system for university library proposed in the study has the advantages of high coverage, low resource consumption, high accuracy and so on, which can provide readers with more accurate and personalized recommendation services.