Design And Implementation Of Personalization Service System Based On Web And Data Mining

The turning from "Information means revolution" to "Information content revolution" has triggered a worldwide wave of building digital library. Digital library will be an information service-based system which basic function is service. It should not only satisfy user's common request, but also provide special resource and service according to different users' type and request. Personalization information service is one of the ways to help it to be an intelligentized knowledge network.This personalization service system described in this paper can give registered users abilities to customize library's digital resource, and provide personalization service for users by using association rules mining algorithm. It mines transaction database of customized digital resources, and discovers mode of customized content, which are association rules. System can give registered users personalization service and recommendation on the basis of association knowledge. Also, with registered users' access history statistic, it can direct librarian what kind of resource to collect.The paper introduces the design and implementation of library personalization service system of Beijing University of Technology in detail. It introduces some main dependable technologies at present, then analyses, compares and selects different methods of these technologies. The paper detailly discusses some crucial technologies of system implementation. It stresses the FP-growth algorithm which is presented by Processor Jiawei Han, discusses its theory, merits and implementation. Also the paper discusses the aspects of system expansibility and efficiency. Finally, the paper demonstrates the use of this system, and gives ideas of system improvement, anticipates the prospect of personalization service.