Analysis and research on the Algorithm of university book recommendation

This paper analyzes the performance of user-based collaborative filtering algorithm and item-based collaborative filtering algorithm in university library lending data and the applicable environment of different algorithms through accuracy and recall rate. The borrowing data of university library are processed by different methods and the processed data are run by different algorithms. The running results of different cutting ratios of training sets and test sets, different recommended quantities and different data processing are recorded. And then look for variables that are relevant. Finally, the best running environment of different recommendation algorithms is found by comparing the data.