Improvement of sub-table making method for extracting pseudo-generalized dynamic reducts in rough set theory

Abstract

In this paper, we discuss an improvement of sub-table making method in the extraction method of pseudo-generalized dynamic reduct (pGDR) in rough set theory. We have proposed and implemented a method for extracting pGDR from large-scale data. This method divides a large-scale decision table into multiple sub-tables, but if the resulting sub-tables are too small, it is difficult to extract pGDRs that reflect the overall characteristics of the original decision tables. In order to solve this problem, we attempt to improve the sub-tables making method.