基于集值信息系统的改进粗集模型

Improvement of Rough Set Model Based on Set Valued Information System

提出了一种新的带有参数的改进容许关系,并讨论了基于该容许关系的粗糙集扩充模型,证明了相关定理。理论分析表明,通过合理设置参数值,可使新扩充模型比以往模型具有更好的分类效果,从而提高了对数据预测、分类的准确度。实例分析进一步验证了改进容许关系及其扩充模型的优越性。

This paper puts forward an improved tolerance relationship which has parameters,discusses the new expanded rough set model based on the relationship,and proves the correlative theorems of the new model. Theoretical analysis shows that the new model,by setting parameters reasonably,has better classification ability than other rough set models. The new model enhances the accuracy of data predication and data classification. The case analysis further validates the advantages of the improved tolerance relationship and the expanded rough set model based on it.