摘要
为得到更高的分类精度和效率,提出了基于一个新的类的关联分类算法CACA(a new class based associative classifica-tion approach)。该方法使用基于策略的类来削减频繁模式的搜索空间;设计一个OR-Tree(ordered rule-tree)的有序规则树来存储规则和他们的信息并且重新定义一个紧凑集,以便构造的分类器也是紧凑唯一的;同步规则的生成和分类器的构造以缩小规则的挖掘空间以便加快规则的生成。实验结果表明,CACA算法在关联分类方法中具有更高的准确度和效率。
To obtain higher classification accuracy and efficiency,a associative classification approach（CACA） based on a new class is proposed.The proposed algorithm use the class based strategic to cut down the searching space of frequent pattern.A structure call Ordered Rule-Tree is designed to store the rules and their information which may also prepare for the synchronization of the two steps;redefine the compact set so that the compact classifier is unique and not sensitive to the rule reduction Experimental result suggested that CACA is making better performances in accuracy and efficiency in Associative classification approaches.
出处
《计算机工程与设计》
CSCD
北大核心
2011年第4期1319-1321,1325,共4页
Computer Engineering and Design
关键词
分类
关联规则
频繁项集
数据挖掘
规则修剪
classification
association rule
frequent item
data mining
rule pruning
