一种垂直分布环境下的特征选择及规则提取算法

An Algorithm for Feature Selection and Rule Extraction in Vertically Partitioned Environment

特征选择及规则提取是数据挖掘过程中的重要环节。Rough集理论提供了一种新的属性约简即特征选择及规则提取工具,但目前Rough集理论研究主要针对单个决策表(或信息系统),分布式环境下的粗糙集理论研究还不多见。文章提出一种垂直分布环境下的特征选择及规则提取算法,算法分析结果表明,该种算法是有效可行的。

Feature selection and rule extraction are the important parts for data mining, rough set theory is a new tool for attribute reduction namely feature selection and nile extraction, however, at present, research on rough set theory aims mainly at a single decision table, very little work has been done in distributed environment. In this paper, we present an algorithm for feature selection and nile extraction in vertically partitioned environment. Algorithm analysis results show the algorithm of this paper is effective and efficient.