基于最大值约束的模糊关联规则挖掘  

Based on Maximum Constraint of Fuzzy Association Rules Mining

作　　者：常浩[1] 李雪梅[2] 

机构地区：[1]太原大学计算机系,太原030009 [2]中北大学电子与计算机科学技术学院,太原030015

出　　处：《电脑开发与应用》2012年第5期7-9,12,共4页Computer Development & Applications

摘　　要：数据挖掘是从数据库中发现潜在有用知识或者感兴趣模式的过程。在数据挖掘领域中主要集中于单一支持度下的关联规则挖掘,在事务数据库中发现项目之间的关联性,而在实际应用中,项目可以有不同的最小支持度,不同的项目可能具有不同的标准去判断其重要性,因此提出一个在最大值支持度约束下,发现有用的模糊关联规则挖掘算法,在该约束下,利用逐层搜索的迭代方法发现频繁项目集,通过实例证明了该挖掘算法是易于理解和有意义的,具有很好的效率。Data mining is used to find the process of potentially useful knowledge and interesting patterns from database. In the field of data mining, it is focused on association rule mining of a single support, founding the correlation between projects from transaction databases. But in real applications, item can have different minimum support and different items may have different criteria to judge its importance. Therefore, based on the maximum support constraints of fuzzy mining algorithm for find useful fuzzy association rules is proposed. Under the constraint, it is the iterative search using level by level method to find frequent item sets. An example is given to demonstrate that the proposed mining algorithm is easy to understand and meaningful, with good efficiency.

关 键 词：数据挖掘 关联规则 最小支持度 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]