Data Mining In Educational Administration Management System

A so-called data warehouse is a collection of subject-oriented time-varying data that is used to support management decision-making process; integration and stability are two basic requirements for data warehouses. Data mining is the process of acquiring knowledge through an analysis of the data stored in database or data warehouse with the aid of artificial intelligence methodology. The combination of data warehouse and data mining techniques can efficiently provide scientific basis for the administrative personnel in enterprises or institutions to make correct decisions.The majority of previous educational administration management systems are OLTP-based, which lack the abilities to support decision making through a comprehensive analysis of the available data; this is especially the case where relevant knowledge is implied in the sea of historical data. As evaluating the effect of teaching administration is an important issue in evaluating the quality of teaching, it is indispensable to conduct a multi-level, multiple-angle analysis of the effect of current teaching administration pattern starting with historical educational administration data and employing data mining techniques. Teaching decision with the aid of the mined rules is a critical requirement for guaranteeing teaching quality and improving students'quality.This thesis address the issue of how to apply the data mining techniques to the educational administration management systems. The main work of this thesis is as follows.①The association-rule-based methodology for data mining is reviewed. Based on a deep understanding of the classical Apriori algorithm, a new algorithm for learning association rules is proposed.②After understanding the features of educational administration management system, we choose to study the problem of how to arrange the courses reasonably through mining students'historical score data. In our study, we use as our mining algorithm the new association rule learning algorithm, and choose as our mining tool the SQL Server Business Intelligence Development Studio toolbox in SQL Server 2005.③Taking as the object to be mined the historical score data for the students majoring in Oil Engineering, we explain the full mining process in detail. Finally, we obtain the reasonable timetable for this specialty and an appropriate criterion for making the teaching plan.