Based On DELPHI And Data Mining Of Student Management And Analysis System

As the sharp increasing numbers of the enrolled students, there are lots of problems which can influence the quality of teaching in the universities. The Data Mining is a procedure of distilling available information and knowledge from mass, incomplete and random data. The Data Mining had been applied and studied in these years, and it has been applied in many domains successfully, such as business, finance and medical treatment. However, little is applied in teaching.Along with the continuous expansion of enrollment among China's higher education institutes, issues like student information management become increasingly significant in that it may influence the school management efficiency as a whole to a great extent. Under such context, Data Mining, based on its previous successfully adoption within business, finance and medical domains, may be a competent program to distill information from mass, random and incomplete data.This paper establishes the database of Geophysics and Information Technology College through SQL Server 2005, with the data covering the information of students enrolled in Geophysics and Information Technology College from 2003 to 2009, namely, including 6 grades, 36 classes, and almost 1200 students'basic information and around 70,000 score information, as well as employment status from 2007 to 2009. It establishes student information table, student score table, class table and course table, and build up data warehouse through SSIS (SQL Server Integration Services). It also visualizes interface through Delphi, composes codes by Pascal, uses ADO database control to carry out inserting, modifying, deleting and updating students'information through SQL. With its friendly interface and fast querying, it can provide all the functions of mainstream student information management system.Based on this data base, a proposal utilizing the technology of Decision Tree to explore the invisible information of students'scores, is raised, with detailed analysis of the specific steps within such process, of the main algorithm and basic rationale and the objective contrast among different algorithms. All these ensure the possibility to spot the present incomplete analysis of students'scores. Added by the collected and pretreated data coming from questionnaires, through C4.5arithmetic, this Decision Tree finally comes into being, as well as corresponding classification criteria.This paper takes example from the institute where the researcher works, raises this project aiming at exploring the hidden valuable information behind students'scores through the application of Decision Tree, thus providing basis for further decision-making during school administrative and teaching practice.