The Associative Classification approach combining the classification and association rule mining techniques is now becoming reliable classification strategy for obtaining moderate sized classifiers with enhanced accuracy rate. The associative classification techniques when applied on large databases produce huge number of rules and make difficult to achieve effective classification model. To overcome the problems underlying the associative classification, this paper focuses on rule ranking criteria used during rule generation process and proposes the unique rule pruning procedures followed by discussing their impact on classification accuracy rate. The goal of the study is to effectively apply the rules produced classify the data and to obtain reduced size classifiers with high classification accuracy rate.

结合了分类和关联规则挖掘技术的关联分类方法现在正成为可靠的分类策略，用于获得具有较高准确率的中等大小的分类器。当关联分类技术应用于大型数据库时，会产生大量规则，并且难以实现有效的分类模型。为了克服关联分类的潜在问题，本文重点介绍规则生成过程中使用的规则排名标准，并提出独特的规则修剪程序，然后讨论它们对分类准确率的影响。该研究的目的是有效地应用所产生的规则对数据进行分类，并获得具有较高分类准确率的尺寸减小的分类器。