Research On Accurate Recommendation Of Library Books Based On User Portrait

As an information sharing center,university library includes almost all languages and all kinds of carrier information.How to make effective use of information resources and give full play to the service role of the library is very important.Nowadays,with the rapid development of Internet,big data and cloud computing,the undifferentiated recommended content of the existing library management system can no longer meet the diversified and personalized needs of users;For the large amount of user data accumulated all year round in the library management system,the data value also needs to be excavated.Therefore,using data technology for innovation has become a new driving force for the development and reform of libraries.It is a tool to process the user’s feature vector and get the user’s feature vector through user’s image.Applying the user portrait to the library field,constructing the portrait model through data analysis,and using the portrait to effectively predict the user preference,user interest and user behavior can realize the accurate recommendation of library books and meet the personalized needs of readers.Firstly,based on the research background of user portrait in the field of library,this paper combs the methods and technical means used by scholars at home and abroad to construct user portrait,and expounds the theoretical and technical concepts related to the construction of user portrait model,so as to provide theoretical support for the subsequent construction of user portrait model suitable for library.Secondly,aiming at the unique application scenario of the library,this paper constructs a user portrait model suitable for the library from data collection and analysis,label extraction to portrait mining,and expounds in detail the technical methods used in the process of forming the portrait.Then,according to the user portrait model,based on the portrait classification,a collaborative filtering recommendation algorithm based on user clustering is proposed for the library.Finally,the actual behavior data of users in Yanshan University Library are selected for example verification and analysis,in order to realize the accurate recommendation of book resources.According to the example verification analysis,the constructed library user portrait model and the collaborative filtering recommendation algorithm based on user clustering are feasible.The research results have made a beneficial exploration for the research on accurate recommendation of library books in the future.The library administrator can outline the fine portrait of readers according to the user data generated by readers in the interaction process of library system,so as to make accurate recommendation of books and improve users’ sense of experience.