TY - JOUR
T1 - Using single axioms to characterize L-rough approximate operators with respect to various types of L-relations
AU - Pang, Bin
AU - Mi, Ju Sheng
N1 - Publisher Copyright:
© 2020, Springer-Verlag GmbH Germany, part of Springer Nature.
PY - 2020/5/1
Y1 - 2020/5/1
N2 - Considering L being a complete Heyting algebra, this paper mainly proposes a general framework of L-rough approximate operators in which constructive and axiomatic approaches are used. In the constructive approach, upper and lower L-rough approximate operators are introduced and their connections with L-relations are investigated. In the axiomatic approach, various types of set-theoretic L-operators are defined. It is shown that each type of L-rough approximate operators corresponding to special kind of L-relations, including serial, reflexive, symmetric, transitive, mediate, Euclidean and adjoint L-relations as well as their compositions, can be characterized by single axioms.
AB - Considering L being a complete Heyting algebra, this paper mainly proposes a general framework of L-rough approximate operators in which constructive and axiomatic approaches are used. In the constructive approach, upper and lower L-rough approximate operators are introduced and their connections with L-relations are investigated. In the axiomatic approach, various types of set-theoretic L-operators are defined. It is shown that each type of L-rough approximate operators corresponding to special kind of L-relations, including serial, reflexive, symmetric, transitive, mediate, Euclidean and adjoint L-relations as well as their compositions, can be characterized by single axioms.
KW - Fuzzy approximate operator
KW - Fuzzy relation
KW - Fuzzy rough set
KW - Fuzzy set
UR - http://www.scopus.com/inward/record.url?scp=85077431362&partnerID=8YFLogxK
U2 - 10.1007/s13042-019-01051-z
DO - 10.1007/s13042-019-01051-z
M3 - Article
AN - SCOPUS:85077431362
SN - 1868-8071
VL - 11
SP - 1061
EP - 1082
JO - International Journal of Machine Learning and Cybernetics
JF - International Journal of Machine Learning and Cybernetics
IS - 5
ER -