粗糙集方法及其在化学模式分类规则挖掘中的应用

简要介绍了粗糙集的基本概念，决策系统的约简步骤和分类规则的挖掘原理，提出了基于信息熵的数据离散化方法，使之充分结合粗糙集特性，具有良好的推广性。又以经典的橄榄油产地判别为例，采用粗糙集方法，无需先验知识，不用设定参数，即能消除冗余的属性和属性值，约简化学系统，从样本数据中挖掘出简明直接、易于理解的产生式分类规则，构建专业意义明确的化学模式分类模型，其预报性能良好，效果令人满意。

Rough Sets and Its Application in Mining Chemical Pattern Classification Rules

The basic concepts of rough sets are introduced briefly. For rough sets can only deal with discrete data,the discretization of continuous data is the key factor in the rough sets applied in chemical domain,we present a method of Chi-Merge discretization based on entropy of information which combined with the characteristic of rough sets,its generalization is well. In problem of the origin discrimination of olive oil,without any additional prior model assumption,rough sets data analysis can eliminate the redundancy of attributes and its value,identify the dependence in the attributes,and we get a collective production rules about the chemical pattern classification system from sample data. When the model of chemical pattern classification is built by these rules,its meaning is very understandable in chemical domain,and the prediction of the model is also well.