Skip to content

Finding a short and accurate decision rule in disjunctive normal form by exhaustive search

Notifications You must be signed in to change notification settings

cebarboza/Explore

 
 

Repository files navigation

Package overview

Functions to find a short and accurate decision rule in disjunctive normal form using the Exhaustive Procedure for LOgic-Rule Extraction (EXPLORE) algorithm. The application performs and exhaustive search on all Boolean Normal Form decision rules.

Package installation

You can install the latest version of EXPLORE like so:

install.packages("remotes")
remotes::install_github("mi-erasmusmc/EXPLORE")

Additional instructions: to be added.

Example usage using iris dataset

library(Explore)
library(farff)

Load data:

data <- farff::readARFF(system.file("examples/iris.arff", package = "Explore"))
output_path <- file.path(getwd(), "output//")

Fit model with defaults and/or input parameters:

model <- Explore::trainExplore(output_path = output_path, 
                               file_name = "iris", 
                               train_data = data, 
                               ClassFeature = "'class'", 
                               PositiveClass = '"Iris-versicolor"')

Predict:

prediction <- Explore::predictExplore(model, test_data = data)

Additional documentation includes: - Vignette code examples in combination with PLP: to be added - Package manual: to be added

Development status

EXPLORE is under active development.

Publication

Rijnbeek, P.R., Kors, J.A. Finding a short and accurate decision rule in disjunctive normal form by exhaustive search. Machine Learning 80, 33–62 (2010). https://doi.org/10.1007/s10994-010-5168-9

About

Finding a short and accurate decision rule in disjunctive normal form by exhaustive search

Resources

Stars

Watchers

Forks

Packages

No packages published

Languages

  • C++ 95.6%
  • R 2.6%
  • Other 1.8%