A New Approach to Hybrid Condition Attribute Reduction Based on Rough Set

We focus on hybrid condition attribute reduction based on rough set. Generally, the process of attribute reduction from a large information system is time consuming. Since its computational complexity increases exponentially with the number of input variables and in multiplication with the size of data patterns, we develop a new approach to attribute reduction by using rough set to deal with the problem. In contrast to traditional attribute reduction, we take advantage of the reduction of the scale of the boundary region of the elementary sets induced by decision attributes. Finally, a example is presented to examine the approach and is derived a sound result.