zislam / ForestPA

Class implementing decision forest algorithm Forest PA, using bootstrap samples and penalized attributes. Uses and depends on SimpleCart.

ForestPA

Class implementing decision forest algorithm Forest PA, using bootstrap samples and penalized attributes. Uses and depends on SimpleCart.

For more information, see:

Adnan, Md Nasim, Md Zahidul Islam (2017). Forest PA: Constructing a Decision Forest by Penalizing Attributes used in Previous Trees. Expert Systems with Applications.

BibTeX:

\article{adnan2017forest,
  title={Forest PA: Constructing a Decision Forest by Penalizing Attributes used in Previous Trees},
  author={Adnan, Md Nasim and Islam, Md Zahidul},
  journal={Expert Systems with Applications},
  year={2017},
  publisher={Elsevier}
}

Installation

Either download ForestPA from the Weka package manager, or download the latest release from the "Releases" section on the sidebar of Github. A video of how to install and use the package can be found here

Compilation / Development

Set up a project in your IDE of choice, including weka.jar and simpleCART.jar as compile-time libraries. simpleCART.jar is available in the Weka package manager.

Valid options are:

-T <num trees> Number of trees requested in the forest. (default 10)

-M <min no> The minimal number of instances at the terminal nodes of each SimpleCart. (default 2)

-N <num folds> The number of folds used in the minimal cost-complexity pruning for each SimpleCart. (default 5)