Improved Algorithm and Further Research of β-reduct for Variable Precision Rough Set

In variable precision rough set, by introducing inclusion degree beta, attribute reduction has been diversified. For example, in variable precision rough set, many researchers have studied beta-reduct, beta lower distribution reduct and beta upper distribution reduct and so on. In this paper, we further research on beta-reduct. We first discuss the monotonic characteristics of beta lower and beta upper approximation with the changes of the parameter beta. Then we cite an example to illustrate the fact that with the change of the inclusion degree beta, there is no monotonic characteristic in beta-reduct. Finally, we develop an improved algorithm of finding beta-reduct by considering the combinations among the condition equivalence classes.