Research On Algorithm And Model Of Book Recommendation System In Universities

University libraries are one of the most important academic places in universities.Millions of books have brought abundant resources for the study and research of teachers and students,but also caused the phenomenon of information overload in university libraries,that is,when facing a large number of books,it is often difficult for teachers and students to find the books they really need.Therefore,the development of personalized book recommendation service for universities is an inevitable trend in the future.On the basis of having consulted relevant literature of China and foreign countries,in this thesis,a hybrid book recommendation system model of universities based on collaborative filtering algorithm and Apriori algorithm is proposed in this thesis,and the specific research process is as follows:Firstly,in order to solve the problem that the scoring data of library users is sparse,the idea of recommending books only among users of the same school is adopted.Meanwhile the same type of books are merged by using the Chinese Library Classification,and the user-the class of book scoring matrix is constructed.Secondly,the user similarity calculation formula based on book borrowing records is improved in this thesis,which increased the accuracy and objectivity of the calculation results,and an improved user-based collaborative filtering algorithm is proposed by integrating the users' natural attribute information.Thirdly,based on the results of Top-N recommendation obtained by improved user-based collaborative filtering algorithm,Apriori algorithm is used in this thesis to mine strong rules in experimental dataset,and these strong rules are used to find highly associated books for Top-N recommendation results.Then the Top-Nrecommendation results and the highly asscioated books are recommended hybridly.Finally,in terms of model construction and effect validation,the hybrid recommendation algorithm is used to construct a hybrid book recommendation system model of universities and the effect of 3 different models of book recommendation system is validated through comparative experiments,which proved the validity of the hybrid book recommendation system model of universities.