Quick algorithm for dominant classes from ordered information systems

摘要

In rough set theory, the ordered information system was introduced to deal with the task of property with preference information in applications. In the ordered information systems, one of the key concepts is the dominant class induced by the dominance relationship. In view of that most of data excavating approaches based on dominance relationship need to compute the dominant class frequently, it is desirable to compute the dominant class in a more efficient way. In this paper, to overcome the limitation of high time-consuming in computing dominant class, a quick algorithm for the computation of dominant class was proposed. The experiments showed that, the new algorithm can reduce the computational time effectively, and is more efficient than existing algorithms.