Ensemble Methods in Ordinal Classification

This repository contains the implementation of the various algorithms presented in Ensemble Methods in Ordinal Data Classification for Weka 3.7.

Included Algorithms

• oAdaBoost: oAdaBoost: An AdaBoost variant for Ordinal Data Classification
• AdaBoost.OR: Combining ordinal preferences by boosting
• AdaBoost.M1w: How to make AdaBoost.M1 work for weak base classifiers by changing only one line of the code
• oDT: Ensemble Methods in Ordinal Data Classification
• Ordinal Random Forests: Ensemble Methods in Ordinal Data Classification

Instalation

Download (Recommended)

1. Download the pre-compiled package from https://github.com/JD557/weka-emiodc/releases
2. Open Weka 3.7 and choose Tools > Package Manager
3. Click on File/URL (under Unofficial) and choose the downloaded .zip
4. Restart Weka
5. You should now an OrdinalEnsembleMethods package
6. The new classifiers should now be available

Compile from source

1. Run sbt package
2. Compress the compiled jar (target/scala-2.10/emiodc_2.10-1.0.jar) alongside the Description.props in a zip
3. Open Weka 3.7 and choose Tools > Package Manager
4. Click on File/URL (under Unofficial) and choose the generated .zip
5. Restart Weka
6. You should now an OrdinalEnsembleMethods package
7. The new classifiers should now be available