Design And Implementation Of Course Work Submission And Management System Based On Web

Along with the development of information technology and network technology,the library digital resources are more and more rich and complete. A typical digitallibrary not only is able to provide information resources to readers, but also shouldprovide personalized information service to different users. And then it can improve theservice quality and service level, improve the utilization rate of information resources.This thesis presents a Web based digital library personalized service mechanism base onthis scenario, and also deliveries a research systematically from an algorithmperspective.This thesis solves the problem of insufficient library personalized informationservice leverage the solution of web mining technology in the domain of data miningbased on the analysis of the connotation, the characteristics, and the way whileproviding the personalized service. Firstly, the thesis presents two Web basedpersonalized service mining association rule discovery technology and clusteringmethod, and analyzes the performance, advantages and disadvantages of variousalgorithm, proposes a algorithm of hierarchical clustering and clustering two clusteringcombination to avoid the defect of the K-means clustering algorithm must first specifythe cluster number k value. Secondly, the thesis proposes the idea of separate design foronline and offline mix with the personalized recommendation based on theimplementation process of the system functionalities. And also, the thesis finallyproposes the user interested model after preparing data, analyzing user interestingbehavior, designing online recommendation engines produce results recommendationset of each module function, and providing the detailed design to the offline part. Thenthe thesis analyzes the content of user browsing and the behavior of user surfing,combines the two analysis results to create user interesting profile, and associated withinformation resource database to create the user interest model database, and representby the two layers structure based on vector space model. The thesis proposes deeplyanalysis and discussion to the algorithms of user interesting model database. Finallyimplementation of the personalized recommendation system includes3major moduleswhich are data preprocessing module, frequent access patterns discovery module,on-line recommendation module and code implementation. The system passed thetesting. The result shows that the system has achieved the users’ demand, and theoverall performance of system achieved the desired design objectives.The thesis proposed the solution by over comes the problems within thepersonalized information service, combines the technologies in the construction of digital library, design and minding the user interested model. It meets the need of thedevelopment for the time been, meets the need of clients, and better increase theutilization rate of Library resources.