Construction Of Library Reader Community Based On Data Mining

The rapid development of digital reading has brought great influence to the library circle.Under this background,the library's reading service and promotion are also facing severe challenges.At present,many libraries still use recommended bibliography,lectures and exhibitions as their reader services.How to reposition and retain readers in Libraries in the new era has triggered a new round of thinking in the industry.Consistently,libraries need to strengthen reading promotion to increase user viscosities and tap new information needs of readers to promote reader reflux.Therefore,making full use of all kinds of user data and service data,building a reader community based on data mining technology to change the passive service mode,can promote the innovation of the traditional library's reader service mode,and can expand the service space of the library.Based on the research and analysis of users' reading needs and library's promotion business needs,this paper determines the design objectives of the reader community platform,and completes the design of the overall system architecture,functional modules,database model,interface design,etc.The reader and administrator complete the detailed functional structure design of the front desk and management end according to the role analysis.This paper focuses on the key technologies and difficulties in the implementation of personalized recommendation.Through in-depth analysis of Library users,book features and mainstream recommendation algorithms,a hybrid community recommendation algorithm based on association rules and collaborative filtering is proposed,which realizes the recommendation function of the platform.The algorithm optimizes the recommendation result set generated by collaborative filtering through association mining.Experiments show that the proposed algorithm is more effective for recommendation.Based on the above requirement analysis and recommendation algorithm,the development and deployment of library reader community platform is realized by using Java + MySQL + Apache architecture.The platform includes several functional modules,such as personalized recommendation,book search,Book scoring and sharing,book-friend activities,and book-friend interaction,to meet the interactive and personalized service needs of readers.By introducing Web Service and Ajax technology,the stability and expansibility of system interaction are further guaranteed.Finally,through functional testing and performance testing,the availability and reliability of the system are verified,and the feasibility of the platform scheme is proved,which provides guidance suggestions for the construction of the library community platform.