【作者】 陈建军；

【导师】 高玉斌；

【作者基本信息】 中北大学 ， 应用数学， 2008， 硕士

【摘要】 随着数据库应用的不断深化,数据库的规模急剧膨胀,人们需要对这些数据进行分析,从中发现有价值的信息。数据挖掘已经成为机器学习、人工智能、数据库等领域的研究热点。它包含关联规则挖掘、预测、分类、聚类、演化分析等多种技术手段。其中关联规则挖掘是一种主要的,也是用途最广的数据挖掘方法。本文即对数据挖掘中的关联规则进行系统研究,深入分析了关联规则的传统支持度-置信度框架、相关度和有效度等衡量标准,并在此基础上将T检验思想引入到了关联规则的衡量中,提出了一种新的关联规则衡量标准-影响度。在对多媒体教学评价现状和相关理论分析的基础上,作者提出了多媒体教学评价的基本原则和多媒体教学评价数据分析的目的,指出了本研究课题中关联规则在多媒体教学评价数据分析中的应用方向。本文将关联规则衡量标准-影响度应用到多媒体教学质量评价数据分析中。采用J2EE的体系结构,用Webwork+Spring+Hibernate架构设计并开发了一个多层的教学评价数据挖掘系统,有效地提高系统的可靠性、可扩展性、可重用性和可维护性。并利用该系统对多媒体教学质量评价数据进行了分析,系统运行结果表明,利用将影响度作为关联规则的衡量标准寻找多媒体教学评价数据中潜在的关联性是可行的、有价值的,可以有效的克服现有衡量标准的一些不足,减少冗余规则的产生。更多还原

【Abstract】 With the deepening of the application of database, the size of database expands quickly, people need to analyze these data and find the worthy information. Consequently data mining has become a research area with increasing importance. It includes lots of measures such as association rules mining, classification and prediction, clustering analysis and evolvement analysis. The main technique among the data mining measures is the association rules mining, which is also the most widely used data mining measure.In this paper, association rule mining was studied, researched and analyzed deep.And the author analyze and discuss the support-confidence framework, the correlativity and the validity. With introducing T-Testing, the Effect as a new evaluation criterion for association rules is proposed.The author analyze the current situation of Multi-media Teaching Evaluation and the related theory. Basic principles and data analysis purposes of Multi-media Teaching Evaluation is proposed in this paper. Following the author provide applications of the association rules in the analysis of Multi-media Teaching evaluation data in this research. In this paper,the association rules algorithm is introduced in the analysis of Multi-media Teaching evaluation data. A multi-layer data mining system based on the popular J2EE framework is presented. The system is modeled and developed separately using Webwork + Spring + Hibernate frame, which improve the maintainability, the reusability and the extendibility. The system is introduced in the analysis of multi-media teaching evaluation data. The result shows that introducing effect based on common approach to association rules mining for the analysis of Multi-media Teaching evaluation data is feasible and valuable and can not only effectively overcome the shortage of the existing evaluation criterion for association rules but also reduce the creation of redundant rules.更多还原