Composite Rough Sets

摘要

There are multiple kinds of data in information systems, e.g., categorical data, numerical data, set-valued data, interval-valued data and missing data. Such information systems are called as composite information systems in this paper. To process such data, composite rough sets are introduced, composite relation is defined and composite classes are used to drive approximations from composite information systems. Lower and upper approximations of a concept are the basis for rule acquisition and attribute reduction in rough set theory. To intuitively compute the approximations, positive, boundary and negative regions, matrix-based method is presented in composite rough sets. A case study validates the feasibility of the proposed method.