VPRS-Based Knowledge Discovery Approach in Incomplete Information System

Abstract

Through changing the equivalence relation in the incomplete information system, a new variable precision rough set model and an approach for knowledge reduction are proposed. To overcome no monotonic property of the lower approximation, a cumulative variable precision rough set model is explored, and the basic properties of cumulative lower and upper approximation operators are investigated. The example proves that the cumulative variable precision rough set model has wide range of applications and better result than variable precision rough set model.