**weka.classifiers.meta**

A regression scheme that employs any classifier on a copy of the data th= at has the class attribute discretized. The predicted value is the expected= value of the mean class value for each discretized interval (based on the = predicted probabilities for each interval). This class now also supports co= nditional density estimation by building a univariate density estimator fro= m the target values in the training data, weighted by the class probabiliti= es.

=20For more information on this process, see

=20Eibe Frank, Remco R. Bouckaert: Conditional Density Estimation with Clas= s Probability Estimators. In: First Asian Conference on Machine Learning, B= erlin, 65-81, 2009.

=20The table below describes the options available for RegressionByDiscreti= zation.

=20
=20
=20
=20
=20
=20
=20
=20
=20
=20
=20
=20
=20
=20
=20
=20
=20
=20
=20
=20
=20
=20
=20
=20
=20
=20
=20
=
=20
=20
=20
=20
=20
=20

=20
Option | Description |
---|---|

classifier | The base classifier to be used. |

debug | If set to true, classifier may output additi= onal info to the console. |

deleteEmptyBins | Whether to delete empty bins after discretiz= ation. |

estimatorType | The density estimator to use. |

minimizeAbsoluteError | Whether to minimize absolute error. =20 |

numBins | Number of bins for discretization. |

useEqualFrequency | If set to true, equal-frequency binning will= be used instead of equal-width binning. |

The table below describes the capabilities of RegressionByDiscretization= .

=20
=20
=20
=20
=20
=20
=20
=20
=20
=20
=20
=20
=20
=20
=20
=20
=20
=20

Capability | Supported |
---|---|

Class | Date class, Missing class values, Numeric cl= ass |

Attributes | Numeric attributes, Nominal attributes, Empt= y nominal attributes, Date attributes, Binary attributes, Missing values, U= nary attributes |

Min # of instances | 2 |