基于粗糙集和信息增益的决策规则生成

Generating decision rules based on rough set and information gain

信息增益技术可以删除信息量低的属性,决策规则的相关系数分析能对决策规则的准确度和覆盖度进行描述。提出了利用信息增益分析技术对决策表属性进行相关分析,然后进行属性和属性值约简,去除决策表中与决策无关的冗余信息,得出决策规则后,运用相关系数衡量决策规则的精确度。信息增益技术为决策规则的生成提供了一个有效的途径。

The technology of information gain can delete the attributes with less information. The analysis of the related coefficients can describe the certainty and coverage of the decision rules. Firstly, the analysis of information gain technology is used to analyze the rela- tionship between attributes. Secondly, an attribute and attribute value reduction are used to delete the superfluous decision information and get the decision rules in a decision table. Finally, based on analysis of the related coefficients, the decision rules are generated after reduction. Related coefficients are proposed to weight the accuracy of decision rules. The technology of information gain provides an effective way to generate the decision rules.