基于变精度粗糙集决策树的构造算法

The Construction Algorithm of Decision Tree Based on Variable Precision Rough Set

在构造决策树的过程中,分离属性选择的标准直接影响分类的效果.基于变精度粗糙集的理论将属性重要度作为选择分离属性的标准.经实验证明,使用该方法构造的决策树与经典ID3决策树算法相比,具有复杂性低,能有效提高分类效果的优点.

In the process of constructing a decision tree, the criteria of selecting partitional attributes influences the efficiency of classification. Based on variable precision rough set theory, the attribute importance is regarded as the criteria for choosing attributes. The experiments show that compared with the ID3 decision tree, this method is simple and can improve the efficiency of classification.