February 2022
FSTA 2022
14/02/22/10:22 Filed in: Conference participation
Fuzzy Sets Theory and Applications, FSTA, Bratislava, Slovakia, Sep 20-24, 2021
We presented the works "Fuzzy closure relations" (by Emilio Muñoz and other colleagues in the department, Manuel Ojeda-Hernández presented the paper) and "Lower Sugeno-like Integral for Multi-Adjoint FCA" presented by Manuel Ojeda-Aciego which, in addition, had a plenary talk entitled "Galois connections between fuzzy unbalanced structures".
Well organised hybrid conference in which most of the participants were in the venue. Small but nice, and with lots of interesting papers and research ideas to discuss about.
Journal paper accepted
14/02/22/10:08 Filed in: Journal paper
F. Pérez-Gámez, D. López-Rodríguez, P. Cordero, Á. Mora, M. Ojeda-Aciego. Simplifying Implications with Positive and Negative Attributes: a Logic-based Approach. Mathematics 2022. To appear
ABSTRACT Concepts and implications are two facets of the knowledge contained within a binary relation between objects and attributes. Simplification Logic (SL) has proved to be valuable for the study of attribute implications in a concept lattice, a topic of interest in the more general framework of Formal Concept Analysis (FCA). Specifically, SL has become the kernel of automated methods to remove redundancy, or obtain different types of bases of implications. Although originally FCA uses only the positive information contained in the dataset, negative information (explicitly stating that an attribute does not hold) has been proposed by several authors, but without an adequate set of equivalence-preserving rules for simplification. In this work, we propose a mixed simplification logic and a method to automatically remove redundancy in implications, which will serve as a foundational standpoint for automated reasoning methods for this extended framework.
ABSTRACT Concepts and implications are two facets of the knowledge contained within a binary relation between objects and attributes. Simplification Logic (SL) has proved to be valuable for the study of attribute implications in a concept lattice, a topic of interest in the more general framework of Formal Concept Analysis (FCA). Specifically, SL has become the kernel of automated methods to remove redundancy, or obtain different types of bases of implications. Although originally FCA uses only the positive information contained in the dataset, negative information (explicitly stating that an attribute does not hold) has been proposed by several authors, but without an adequate set of equivalence-preserving rules for simplification. In this work, we propose a mixed simplification logic and a method to automatically remove redundancy in implications, which will serve as a foundational standpoint for automated reasoning methods for this extended framework.
Journal paper accepted
10/02/22/08:27 Filed in: Journal paper
P. Cordero, M. Enciso, D. López-Rodríguez, Á. Mora. fcaR, Formal Concept Analysis with R. R Journal, 2022. To appear
ABSTRACT Formal concept analysis (FCA) is a solid mathematical framework to manage information based on logic and lattice theory. It defines two explicit representations of the knowledge present in a dataset as concepts and implications. This paper describes an R package called fcaR that implements FCA's core notions and techniques. Additionally, it implements the extension of FCA to fuzzy datasets and a simplification logic to develop automated reasoning tools. This package is the first to implement FCA techniques in R. Therefore, emphasis has been put on defining classes and methods that could be reusable and extensible by the community. Furthermore, the package incorporates an interface with the arules package, probably the most used package regarding association rules, closely related to FCA. Finally, we show an application of the use of the package to design a recommender system based on logic for diagnosis in neurological pathologies.
Library: https://malaga-fca-group.github.io/fcaR/
ABSTRACT Formal concept analysis (FCA) is a solid mathematical framework to manage information based on logic and lattice theory. It defines two explicit representations of the knowledge present in a dataset as concepts and implications. This paper describes an R package called fcaR that implements FCA's core notions and techniques. Additionally, it implements the extension of FCA to fuzzy datasets and a simplification logic to develop automated reasoning tools. This package is the first to implement FCA techniques in R. Therefore, emphasis has been put on defining classes and methods that could be reusable and extensible by the community. Furthermore, the package incorporates an interface with the arules package, probably the most used package regarding association rules, closely related to FCA. Finally, we show an application of the use of the package to design a recommender system based on logic for diagnosis in neurological pathologies.
Library: https://malaga-fca-group.github.io/fcaR/