On three types of $L$-fuzzy $β$-covering-based rough sets

Abstract: In this paper, we mainly construct three types of $L$-fuzzy $\beta$-covering-based rough set models and study the axiom sets, matrix representations and interdependency of these three pairs of $L$-fuzzy $\beta$-covering-based rough approximation operators. Firstly, we propose three pairs of $L$-fuzzy $\beta$-covering-based rough approximation operators by introducing the concepts such as $\beta$-degree of intersection and $\beta$-subsethood degree, which are generalizations of degree of intersection and subsethood degree, respectively. And then, the axiom set for each of these $L$-fuzzy $\beta$-covering-based rough approximation operator is investigated. Thirdly, we give the matrix representations of three types of $L$-fuzzy $\beta$-covering-based rough approximation operators, which make it valid to calculate the $L$-fuzzy $\beta$-covering-based lower and upper rough approximation operators through operations on matrices. Finally, the interdependency of the three pairs of rough approximation operators based on $L$-fuzzy $\beta$-covering is studied by using the notion of reducible elements and independent elements. In other words, we present the necessary and sufficient conditions under which two $L$-fuzzy $\beta$-coverings can generate the same lower and upper rough approximation operations.