Hitachi Vantara Pentaho Community Wiki
Child pages
  • WrapperSubsetEval
Skip to end of metadata
Go to start of metadata

Package

weka.attributeSelection

Synopsis

WrapperSubsetEval:

Evaluates attribute sets by using a learning scheme. Cross validation is used to estimate the accuracy of the learning scheme for a set of attributes.

For more information see:

Ron Kohavi, George H. John (1997). Wrappers for feature subset selection. Artificial Intelligence. 97(1-2):273-324.

Options

The table below describes the options available for WrapperSubsetEval.

Option

Description

classifier

Classifier to use for estimating the accuracy of subsets

evaluationMeasure

The measure used to evaluate the performance of attribute combinations.

folds

Number of xval folds to use when estimating subset accuracy.

seed

Seed to use for randomly generating xval splits.

threshold

Repeat xval if stdev of mean exceeds this value.

Capabilities

The table below describes the capabilites of WrapperSubsetEval.

Capability

Supported

Class

Date class, Numeric class, Missing class values, Nominal class, Binary class

Attributes

Unary attributes, String attributes, Empty nominal attributes, Relational attributes, Missing values, Numeric attributes, Date attributes, Binary attributes, Nominal attributes

Min # of instances

5

  • No labels