AdditiveRegression

Package

weka.classifiers.meta

Synopsis

Meta classifier that enhances the performance of a regression base classifier. Each iteration fits a model to the residuals left by the classifier on the previous iteration. Prediction is accomplished by adding the predictions of each classifier. Reducing the shrinkage (learning rate) parameter helps prevent overfitting and has a smoothing effect but increases the learning time.

For more information see:

J.H. Friedman (1999). Stochastic Gradient Boosting.

Options

The table below describes the options available for AdditiveRegression.

Option	Description
classifier	The base classifier to be used.
debug	If set to true, classifier may output additional info to the console.
numIterations	The number of iterations to be performed.
shrinkage	Shrinkage rate. Smaller values help prevent overfitting and have a smoothing effect (but increase learning time). Default = 1.0, ie. no shrinkage.

Capabilities

The table below describes the capabilites of AdditiveRegression.

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