Font Size: a A A

Design And Implementation Of Analysis System Of Data-mining-based Campus Card Data

Posted on:2019-02-18Degree:MasterType:Thesis
Country:ChinaCandidate:X K ShiFull Text:PDF
GTID:2417330548460298Subject:Computer technology
Abstract/Summary:PDF Full Text Request
The campus cards in higher vocational colleges has developed for many years,but there lacks an analysis on the characteristics of students' consumption in the practical managements and applications.It is hard to find the inner valuable data information.This paper conducts a deep exploration to establish the analysis system of data-mining-based campus card data with the methods of query-authority classifications of analysis results,campus-card data-integrating managements and related consumption analysis.This paper deeply analyzes the structure and composition of campus-card data system in Xingtai Polytechnic College,adopts the Struts+Spring+Hibernate Framework and MVC Three-layer Architecture based on actual demands of different roles and selects SQL Server database as systematic background to complete the design of whole architecture of the system.The related tables have been built based on the analysis on users,roles,and operation authorities and their relationships from the aspect of user-authority setting and the control of different granularity-degree user-authorities has been realized by controlling the three layers,namely,router,view and request.As for data management,this paper makes a design on dimensional models,fact tables and metadata in the mode of data warehouse.And by means of ETL,these data have been extracted and scheduled from the campus-card database and input into the data warehouse ready for data analysis.Finally,the statistical analysis on data related to campus-card status and consumption situation has been conducted with selection clustering algorithm to aquire the micro and macro information associated with consumption situation;meanwhile,FP-growth algorithm was used to mine the preprocessed data in the impoverished-student identifying practices,and the association rules have been built up accordingly to find out the relationships and rules among the consumption levels,academic performance,family condition and other related elements of impoverished-student group.Actually,the analysis and explorations on the campus-card data conducted by this system is beneficial for the campus management and its decision supports and lays the solid foundation to create the complete analysis platform and realize the decisions of higher-level leadership.
Keywords/Search Tags:data mining, data warehouse, ETL, association rules, FP-growth algorithm
PDF Full Text Request
Related items