Concept lattice simplification with fuzzy linguistic information based on three-way clustering

Numerous linguistically valued facts from the actual world have been modeled using the fuzzy linguistic approach. Concept lattice theory has the potential to be used in information processing in imprecise language environments as a methodology for data analysis and knowledge representation. The concept lattice with linguistic values can manage fuzzy linguistic data which are either comparable or incomparable. However, the acquired conceptual knowledge is difficult to understand due to the substantial amount of linguistic concept knowledge in the concept lattice with imprecise linguistic information. This work proposes a linguistic-valued layered concept lattice simplification method based on three-way clustering to reduce the scale of the linguistic-valued layered concept lattice. In order to create the linguistically valued layered concept lattice, a reconstruction function is first used to produce an attribute selection model in fuzzy linguistic formal contexts. Second, the intent similarity and extent similarity are employed to achieve the concept similarity measure in linguistic-valued layered concept lattices, taking into account the relationship among various layers of fuzzy linguistic values. To get the initial concept hard clustering results, the linguistic-valued layered concepts are then clustered using k-modes clustering. In addition, we explore the classification of boundary linguistic-valued layered concepts and achieve the three-way clustering findings of linguistic-valued layered concepts to deal with the ambiguity of linguistic expressions. Finally, experimental findings on real -world datasets show how effective the proposed method is for simplifying linguistic-valued layered concept lattices.