摘要
关联规则是数据挖掘的主要技术。文中介绍了关联规则的基本概念,阐述了自顶向下算法的基本思想和存在的不足,扩展了相关定义和性质,提出了基于自顶向下算法基础上的改进算法。该算法的主要特点是运用集合运算的思想和递归的方法,保存前面扫描时比较运算的结果进行最大频繁集的查找。最后用实例进行仿真实验并做了比较分析,效率有所提高。
Association rules are the main technique for data minning. Introduce the basic concept of mining association rules, elaborate the basic thought and existent shortage of the top to bottom algorithm, expand the related definition and property and put forward improved algorithm. Based on top to bottom algorithm, restored last results that are scanning, the max frequent itemset can be refined by using the gathering thought and the recursion method. Finally, carry on analysis by using solid examples. This algorithm is suitable for the large database.
出处
《计算机技术与发展》
2008年第2期136-138,155,共4页
Computer Technology and Development
基金
国家自然科学基金项目(40574002)
广西自然科学基金项目(0448076)
关键词
数据挖掘
关联规则
频繁集
支持数
自顶向下
data mining
association rules
frequent itermset
supporting count
top to bottom
