Analysis And Design Of Library Management System Based On Data Mining

With the development of computer technology, the mode of traditional library management is giving way to that of digitization management. Therefore, offering the readership prompt, accurate and personalized service has become an urgent need in the data era. How to improve service and how to gain authentic and effective data from the system characterized with expanding scale and increasing complexity are the subject that the book administrative staff need to explore. The book administrative section is supposed to manage the online books in an convenient and effective way.According to the original data from the college library management system, this thesis, based on data mining B/S structure, studies the book management system through adopting data mining technology, cluster analysis and association rules algorithm. The system incorporates with J2EE and uses data mining technology to help the readership access the library resources better, which provides a platform for university readership to scan, browse and discuss online the digitization books quickly. The thesis discusses that the application of optimized Apriori algorithm is helpful to reduce transaction set scanning and improve data mining. It also makes cluster analysis come true by using K-means algorithm and understands the readerships’ book-borrowing. The application of association rule discovers the relativity of books and provides technical support for the personalized service, building such functions as data mining application mode, book recommendation and personalized service in the library management system.Data mining from the original college management system provides data support for the library routine management and book purchasing, improves the readerships’ satisfaction for library management and meets the actual demands of the college library management.