Linguistic rough sets

Manish Agarwal, Themis Palpanas

Abstract: We introduce linguistic rough set (LRS) by integrating linguistic quantifiers in the rough set framework. The proposed LRS is inspired by the ways in which humans process imprecise information. It operates directly with the linguistic summaries and caters to imprecision implicit in the real world with partial knowledge. The measures of LRS are developed and its properties are investigated in detail. An approach is proposed for approximation of fuzzy concepts with the proposed LRS. This approach is applied in a real world case-study on the credit scoring analysis problem.