Package

weka.classifiers.meta

Synopsis

Combines several classifiers using the ensemble selection method. For more information, see: Caruana, Rich, Niculescu, Alex, Crew, Geoff, and Ksikes, Alex, Ensemble Selection from Libraries of Models, The International Conference on Machine Learning (ICML'04), 2004. Implemented in Weka by Bob Jung and David Michael.

Available in Weka 3.7.1. Available via the package management system for Weka >= 3.7.2 (ensembleLibrary).

Options

The table below describes the options available for EnsembleSelection.

Option	Description
algorithm	the algorithm used to optimizer the ensemble
debug	If set to true, classifier may output additional info to the console.
greedySortInitialization	Whether sort initialization greedily stops adding models when performance degrades.
hillclimbIterations	The number of hillclimbing iterations for the ensemble selection algorithm.
hillclimbMetric	the metric that will be used to optimizer the chosen ensemble..
library	An ensemble library.
modelRatio	The ratio of library models that will be randomly chosen to be used for each iteration.
numFolds	The number of folds used for cross-validation.
numModelBags	The number of "model bags" used in the ensemble selection algorithm.
replacement	Whether models in the library can be included more than once in an ensemble.
seed	The random number seed to be used.
sortInitializationRatio	The ratio of library models to be used for sort initialization.
validationRatio	The ratio of the training data set that will be reserved for validation.
verboseOutput	Whether metrics are printed for each model.
workingDirectory	The working directory of the ensemble - where trained models will be stored.

Capabilities

The table below describes the capabilites of EnsembleSelection.

Capability	Supported
Class	Nominal class, Numeric class, Binary class
Attributes	Unary attributes, Empty nominal attributes, Nominal attributes, Date attributes, Missing values, Numeric attributes, Binary attributes
Min # of instances	1

Browser not supported