Implementation of the knowledge discovery framework ForEx++, which was published in:
Md Nasim Adnan and Md Zahidul Islam: ForEx++: A New Framework for Knowledge Discovery from Decision Forests In: Australasian Journal of Information Systems Vol 21, 2017.
This algorithm processes a decision forest and provides a list of high-quality rules that account for each class.
@article{adnan2017forex,
title={ForEx++: A New Framework for Knowledge Discovery from Decision Forests},
author={Adnan, Md Nasim and Islam, Md Zahidul},
journal={Australasian Journal of Information Systems (AJIS)},
volume={21},
pages={1--20},
year={2017}
}
Either download ForExPlusPlus from the Weka package manager, or download the latest release from the "Releases" section on the sidebar of Github. A video on the installation and use of the package can be found here.
This repository contains a Netbeans project. Import into Netbeans and include weka.jar, SysFor.jar, and ForestPA.jar as compile-time libraries. SysFor.jar and ForestPA.jar are available in the Weka package manager.
-P
Whether to print the decision forest that the ForEx++ rules were selected from
(default false)
-Z
Whether to remove rules with no coverage before calculating mean coverage,
support, and rule length
(default true)
-GC
Whether to group rules by class value in the final output.
(default true)
-E <acc | cov | len>
Sort Method for Displaying Rules.
(Default = sort by rule accuracy)
-UA
Whether to use accuracy in selecting ForEx++ rules
(default true)
-UC
Whether to use coverage in selecting ForEx++ rules
(default true)
-UR
Whether to use rule length in selecting ForEx++ rules
(default true)