Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.
Comment: Migrated to Confluence 4.0

Package

weka.classifiers.trees

Synopsis

Class for building a best-first decision tree classifier. This class uses binary split for both nominal and numeric attributes. For missing values, the method of 'fractional' instances is used.

For more information, see:

Haijian Shi (2007). Best-first decision tree learning. Hamilton, NZ.

Jerome Friedman, Trevor Hastie, Robert Tibshirani (2000). Additive logistic regression : A statistical view of boosting. Annals of statistics. 28(2):337-407.

Available in Weka 3.6.x - 3.7.1. Available via the package management system for Weka >= 3.7.2 (bestFirstTree).

Options

The table below describes the options available for BFTree.

Option

Description

debug

If set to true, classifier may output additional info to the console.

heuristic

If heuristic search is used for binary split for nominal attributes.

minNumObj

Set minimal number of instances at the terminal nodes.

numFoldsPruning

Number of folds in internal cross-validation.

pruningStrategy

Sets the pruning strategy.

seed

The random number seed to be used.

sizePer

The percentage of the training set size (0-1, 0 not included).

useErrorRate

If error rate is used as error estimate. if not, root mean squared error is used.

useGini

If true the Gini index is used for splitting criterion, otherwise the information is used.

useOneSE

Use the 1SE rule to make pruning decision.

Capabilities

The table below describes the capabilites of BFTree.

Capability

Supported

Class

Binary class, Nominal class

Attributes

Unary attributes, Binary attributes, Empty nominal attributes, Nominal attributes, Missing values, Numeric attributes

Min # of instances

1