Attribute Reduction as Calculation of Focus in Granular Reasoning

Granular reasoning proposed by Murai et al. is a mechanism for reasoning using granular computing, and the concept of "focus" has been proposed as a key concept of granular reasoning. On the other hand, the authors have redefined the concept of focus, and have proposed another concept of granularity, called visibility. However, as a result of the redefinition of focus, calculation of focus is very difficult because we need to check each atomic sentence in the visibility whether the atomic sentence is implied from the set of sentences we use in the current step of reasoning. In this paper, we propose a calculation method of focus in granular reasoning based on attribute reduction of rough set theory.