提出了一种基于决策属性支持度的属性相对约简算法。
A kind of attribute relative reduction for decision attribute support degree was proposed.
针对不一致数据库，定义属性权重及缺省规则加权支持度概念，在此基础上给出一种缺省规则挖掘算法。
The attribute weight and weighted support of default regular are defined by using the conditional entropy and a mining algorithm of default regulars are given for inconsistent database.
将属性值约简和数据挖掘相结合，给出支持度、置信度、覆盖度的定义。
This paper associates attributive value reduction with data mining and proposed three concepts: support, confidence and coverage.
应用推荐