Attribute implications in data with fuzzy attributes using armstrong axioms
01 Aug 2017-
TL;DR: This paper focuses on generating attribute implications in fuzzy setting by using power subsets of fuzzy attributes using Armstrong axioms for reflexivity, augmentation and transitivity.
Abstract: In the last decade researchers have paid more attention to Formal concept analysis (FCA) in the fuzzy setting and its applications. Less attention has been paid for attribute implications in FCA with fuzzy setting and its applications. In this paper we focus on generating attribute implications in fuzzy setting. For this purpose a method is proposed using power subsets of fuzzy attributes. The redundancy is removed using Armstrong axioms (i.e. reflexivity, augmentation and transitivity) with an illustrative example.
Citations
More filters
TL;DR: A rule extraction method is presented to help people make more reasonable decisions, combining the confidence and support degree of linguistic-valued intuitionistic fuzzy decision rules.
12 citations
TL;DR: This paper focuses on introducing a method for bipolar fuzzy attribute implications and its measurement using accuracy function with an illustrative example and aims to help multi-decision process based on user-required chosen attributes.
7 citations
6 citations
TL;DR: This work puts forward a linguistic-valued layered concept lattice for meeting the requirements of different experts at different levels based on lattice implication algebra, and adopts the deleting or uniting strategy to deal with the redundant rules.
Abstract: Formal concept analysis as an effective tool for data analysis and knowledge acquisition can be used to describe the potential relation between objects and attributes. In order to handle linguistic uncertainty information with comparability and incomparability, we propose a kind of linguistic-valued formal concept analysis approach based on lattice implication algebra. Firstly, by setting different linguistic-valued trust degrees, we put forward a linguistic-valued layered concept lattice for meeting the requirements of different experts at different levels. Secondly, the rule extraction algorithm of the linguistic-valued layered concept lattice with the trust degree is given to acquire non-redundant linguistic-valued rules with different trust degrees by using the linguistic-valued weakly consistent formal decision context. Then, aiming at the same premise or conclusion for the different rules, we adopt the deleting or uniting strategy to deal with the redundant rules. The updated and simplified rules can make the rule acquisition easier and the linguistic-valued decision rules extracted are more compact. Finally, the effectiveness and practicability of the proposed approach are illustrated by the comparison analysis.
5 citations