一种基于无或言规则集的改进的关联规则挖掘算法  

An Improved Association Rule Mining Algorithm Based on Disjunction-free Sets

作　　者：卢炎生[1] 王莉[1] 赵栋[1] 

机构地区：[1]华中科技大学计算机学院,武汉430074

出　　处：《计算机工程》2005年第5期99-101,共3页Computer Engineering

摘　　要：提出了一个基于无或言规则集的改进的关联规则算法——IHPD。无或言规则集(disjunction-freesets)是一种精简集表示。这一算法总结了HLinEx,IHP和DHP算法的优点,极大地改善了算法性能,并且使算法的使用不仅仅局限于长类型的频繁项目集挖掘。实验结果表明IHPD算法在性能上比HLinEx更高效。This paper proposes an improved association rule mining algorithm based on disjunction-free sets: IHPD.The representation of disjunction-free sets is a kind of condensed representation. This algorithm combines all the advantages of HLinEx, IHP and DHP, then improves the algorithm performance greatly. Furthermore it breaks the restriction that the algorithm can be efficient only in difficult cases.The experiments show that IHPD can be more efficient than HLinEx.

关 键 词：数据挖掘 精简集 频繁项目集 

分 类 号：TP312[自动化与计算机技术—计算机软件与理论]