SDEFSR: SDEFSR: A package for Subgroup Discovery with Evolutionary...
In SDEFSR: Subgroup Discovery with Evolutionary Fuzzy Systems

Description Details SDEFSR functions Author(s)

The SDEFSR package provide a tool for read KEEL datasets and four evolutionary fuzzy rule-based algorithms for subgroup discovery.

The algorithms provided works with datasets in KEEL, ARFF or CSV format and also with data.frame objects.

The package also provide a Shiny app for making the same tasks that the package can do and can display some additional information about data for making an exploratory analysis.

The algorithms provided are Evolutionary Fuzzy Systems (EFS) which take advantages of evolutionary algorithms for maximize more than one quality measure and fuzzy logic, which makes a representation of numerical variables that are more understandable for humans and more robust to noise.

The algorithms in the SDEFSR package support target variable with more than two values. However, this target variables must be categorical. Thus, if you have a numeric target variable, a discretization must be perfomed before executing the method.

MESDIF Multiobjective Evolutionary Subgroup DIscovery Fuzzy rules (MESDIF) Algorithm.
NMEEF_SD Non-dominated Multi-objective Evolutionary algorithm for Extracting Fuzzy rules in Subgroup Discovery (NMEEF-SD).
read.dataset reads a KEEL, ARFF or CSV format file.
SDIGA Subgroup Discovery Iterative Genetic Algorithm (SDIGA).
SDEFSR_GUI Launch the Shiny app in your browser.
FUGEPSD Fuzzy Genetic Programming-based learning for Subgroup Discovery (FuGePSD) Algorithm.
plot.SDEFSR_Rules Plot the discovered rules by a SDEFSR algorithm.
sort.SDEFSR_Rules Sort the discovered rules by a given quality measure.
SDEFSR_DatasetFromDataFrame Reads a data.frame and create a SDEFSR_Dataset object to be execute by an algorithm of this package.