Design And Research Of A Personalized Book Recommendation System Based On Tags

At present,the development of the active books recommendation service in the university library is lagging behind,the analysis and mining of users' information needs is insufficient.Whether it has already had the target books or not,it will be very waste time to find out some satisfacting books in a resourceful library.This paper uses "tags" to dredge up the user's personal interests and actively recommend books that truly meet the user's needs.From the past "people looking for books" translates to the present "books to find people",it is appropriate to help users quickly find books and efficiently use library resources.The specific research objectives of this paper are as follows:1.In order to remove the obstacle of screening information for new readers and attract more new readers,a large number of books collected in the library are first classified and recommended,and the book classification recommendation is also very beneficial to the expansion of label data.2.There are three kinds of recommendation systems based on user similarity,item similarity and label.This paper not only uses the same or similar labels to judge the books that users need,but also calculates the similarity of users by the same labels.Therefore,the two recommendation methods are closely combined to achieve personalized recommendation of books.3.Readers communicate and interact with the book recommendation system through the user interface.Therefore,the user interface is very clear and lets the reader know how to use the system.The user interface is also very simple,the reader does not need to take the time to understand,thus realizes the good user interface.4.According to the survey,people wait only eight seconds,beyond this time,people will feel impatient.There are more personalized tags produced by users,and the amount of data needed to be handled by the tag-based recommendation system is large.Therefore,the efficiency of the system is very important to the popularization and application of the system.In order to improve the accuracy and user experience,this paper improves the data model and recommendation algorithm respectively.1.The innovative ideas of the data model in this paper are as follows:Metadata,such as book authors,is also used as a social label to solve the problem of sparse data.2.The innovative ideas of the recommends algorithm in this paper are as follows:In the past,researchers have calculated the similarity of users based on the resources they care about.Recommend books for readers according to the similarity.In this paper,we will calculate the similarity of the tags produced by users.If many people choose one or more of the same labels,they will probably have the same preferences.This paper analyzes the students' individualized demand and designs and implements the books personalized recommendation system based on this records from books,student browsing,evaluating,collecting and borrowing books.Recommend personalized books and labels for users so that users can find their own information and improve the satisfaction of users to the library.The development environment of personalized book recommendation system based on tag is eclipse.the data is stored in mysql database.The system mainly includes user registration and login module,personalized recommendation module,book list module,digital resource module,book search module and my library six functional modules.This system adopts the structure of b / s,and users can access the system only by installing a browser,which is beneficial to the popularization of the system and the convenience of the users.