Knowledge Reduction of Interval-Valued Fuzzy Information Systems Based on Variable Precision Tolerance Relation

摘要

The incomplete information systems is described by interval-valued fuzzy information systems in this paper, which are generalized models of single-valued information systems. And a generalized variable precision tolerance relation is defined in interval-valued fuzzy information system by tolerance degree of condition attributes, then the generalized decision attributes is obtained by means of the tolerance relation. Based on the generalized variable precision tolerance relation, the problem of knowledge reduction is investigated and the discernibility matrices of the generalized decision attributes in interval-valued fuzzy decision information systems are studied. Finally, the methods of the knowledge reduction for the interval-valued fuzzy information systems are presented with some examples.