基于数据挖掘技术的高校辅助决策支持系统研究

【作者】 陈善柳；

【导师】 费洪晓；

【作者基本信息】 中南大学 ， 计算机应用技术， 2004， 硕士

【摘要】 数据挖掘研究如何从大量的数据中智能地、自动地提取出有价值的知识和信息,是当前相当活跃的研究领域。近年来,随着我国高等教育事业的飞速发展及教育信息化建设的快速发展,知识的自动获取已经成为制约高校进一步发展的“瓶颈”。因此,研究应用数据挖掘技术,开发适用于高校管理决策的支持系统,具有理论意义和重要的实用价值。 论文系统分析了高校管理决策支持系统的特点,提出了高校建立决策支持系统的途径与方法,即在建立数据仓库(DW)基础上进行数据挖掘(DM),并与在线分析处理(OLAP)技术相结合以提高系统辅助决策能力。同时本文系统归纳总结了现有的数据挖掘技术。简单介绍了一般数据挖掘的过程。 论文详细阐述了数据挖掘在高校管理决策支持系统中的应用过程:数据挖掘目标的确立、数据源的选择、数据准备、处理与评价模型、验证模型、实施与维护模型,特别是对其中的重点数据准备进行了深入的剖析。根据高校管理数据的具体特点对数据清理方法进行了有益的探索与尝试。最后结合SQL Server 2000中的数据转换服务(DTS),阐述了SQL Server数据仓库的建立,并在此基础上进行数据挖掘,据此介绍了建立高校管理决策支持系统的设计与实现过程。 系统模型的开发过程中采用某校2000级学生的数据开发模型而经过重采样后2001级学生的数据验证模型,实验结果表明该模型在2001级学生的数据上同样具有良好的表现,模型是健壮的。整个系统的设计采用执行包的形式,使系统具有很好的模块化特性,便于后期维护和扩展。实践表明目前该系统较好的适应了高校管理信息化的发展步伐,提高了高校管理决策能力和水平。更多还原

【Abstract】 Data Mining, a new generation of tools and techniques for automatic and intelligent database analysis, is an active area with the promise for a high payoff in many business and scientific application. On the other hand, knowledge discovery has been a bottleneck with the rapid development of high school information technology. To deal with this challenge, Data Mining technology is studied and applied to high school decision support systems in this paper.This paper analyzed the feature of high school decision support systems and introduced the method of building high school DSS. The method was building DW firstly, then DM based on which. DM Combined with OLAP to improve the ability of the DSS. This paper make a summary of the current technology of Data Mining and introduced the process of data mining singlely.This paper expounded the application course of dta mining in high school DSS in detail,such as the goal of mining models 、 the data source of data mining、 dealing and evaluating the mining models、 checking the mining models、 carring out and defending the mining models.This paper deeply analysised clearing the data especially. According to the feature of high school management information, this paper explored the method of clearing the data usefully. Finaly, combining Microsoft Data Transformation Services (DTS) in SQL Server 2000,this paper expounded the method of established the data warehouse in SQL Server 2000, then DM based on which. This paper introduced the process of designing and realizing.Some school 2000 grades of students’ data were used to establish the mining model and verify the mining model based on 2001 grades of students’ data after sampling again in this paper. The experimental result indicated that the model have good behavior too on 2001 grades ofstudents1 data, the model was stalwart. The design of the whole system adopted the form of executing package, which make the system have very good module characteristics, maintainability and expansion. Practice indicated that the sytem should be better adaptation university’s manage informationalized developing steps at present and have improved university’s administrative decision ability and level.更多还原