Lfl: An R package for linguistic fuzzy logic

Abstract

This paper presents an R package that supports the use of fuzzy relational calculus and linguistic fuzzy logic in data processing applications. The lfl package enables computing compositions of fuzzy relations enhanced with distinct extensions, such as excluding features, unavoidable features, or generalized quantifiers. Furthermore, it provides tools for transformation of data into fuzzy sets representing linguistic expressions, mining of linguistic fuzzy association rules, and performing an inference on fuzzy rule bases using the Perception-based logical deduction (PbLD). The package also enables to use the Fuzzy rule-based ensemble, a tool for time series forecasting based on an ensemble of forecasts from several individual methods implemented in R. To the best of the authors' knowledge, there is no other open source software that would provide free tools covering the above-described fragments of the fuzzy modeling area. Therefore, we find highly desirable to allow the community to get familiar with the tools as well as their implementation.