Rough set and rough classification based on imperfect information systems

Rough set theory is a relatively new soft computing tool, it can effectively analyze and process information systems. Traditionally, the information system is assumed to be perfect, i.e. attribute values are not missing and supposed to be precise. However, the imperfect information system is ubiquitous. We investigate rough set and rough classification of an imperfect information system (include missing data and imprecise data), which are illustrated with a running example. The approach extends applied categories of Pawlak's (1998) rough set theory.