Original scientific paper

Interpretation of Fuzzy Attribute Subsets in Generalized One-Sided Concept Lattices

Peter Butka ; Faculty of Electrical Engineering and Informatics, Department of Cybernetics and Artificial Intelligence Technical University of Košice, Košice, Slovakia
Jozef Pócs orcid.org/0000-0003-0405-8579 ; Department of Algebra and Geometry, Palacký University Olomouc, Olomouc, Czech Republic and Mathematical Institute, Slovak Academy of Sciences, Košice, Slovakia
Jana Pócsová ; BERG Faculty, Institute of Control and Informatization of Production Processes Technical University of Košice, Košice, Slovakia

Abstract

In this paper we describe possible interpretation and reduction of fuzzy attributes in Generalized One-sided Concept Lattices (GOSCL). This type of concept lattices represent generalization of Formal Concept Analysis (FCA) suitable for analysis of datatables with different types of attributes. FCA as well as generalized one-sided concept lattices represent conceptual data miningmethods. With growing number of attributes the interpretation of fuzzy subsets may become unclear, hence another interpretation of this fuzzy attribute subsets can be valuable. The originality of the presented method is based on the usage of one-sided concept lattices derived from submodels of former object-attribute model by grouping attributes with the same truth value structure. This leads to new method for attribute reduction in GOSCL environment.

Keywords

Generalized one-sided concept lattices; Galois connections; object-attribute model

Hrčak ID:

111879

URI

https://hrcak.srce.hr/111879

Publication date:

10.12.2013.

Visits: 904 *