ACCELERATION METHOD FOR ATTRIBUTE REDUCTION BASED ON THREE-WAY DECISIONS

Attribution reduction is one of the key topics in the field of rough set theory. There is a redundant calculation in the traditional reduction resulting in lots of time consumption. To solve this problem, an acceleration method for attribute reduction based on sequential three-way decisions are proposed. The specific steps are as follows: 1) calculate the significance of the attribute in the decision system; 2) the attributes will be divided into three domains in terms of the significance of the corresponding attribute. And the attribute with maximal significance will be classified into the positive domain. The attributes whose significance value equal zero will be classified into the negative domain and other attributes will be classified into the boundary domain; 3) calculate the significance of the attributes in the boundary domain cyclically and divide the obtained result, until the set of attributes in the positive domain intended constraints is satisfied. In this research, 6 UCI data sets are selected to conduct experiments in the global and local reduction environments, respectively. The experimental results show that the proposed method can effectively reduce the time consumption of searching reduct in such two environments and will not contribute to a poor classification performance.