BFTree (weka-stable 3.6.8 API)

java.lang.Object
- weka.classifiers.Classifier
- - weka.classifiers.RandomizableClassifier
  - - weka.classifiers.trees.BFTree

All Implemented Interfaces:: Serializable, Cloneable, AdditionalMeasureProducer, CapabilitiesHandler, OptionHandler, Randomizable, RevisionHandler, TechnicalInformationHandler

public class BFTree
extends RandomizableClassifier
implements AdditionalMeasureProducer, TechnicalInformationHandler

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.

BibTeX:

 @mastersthesis{Shi2007,
    address = {Hamilton, NZ},
    author = {Haijian Shi},
    note = {COMP594},
    school = {University of Waikato},
    title = {Best-first decision tree learning},
    year = {2007}
 }
 
 @article{Friedman2000,
    author = {Jerome Friedman and Trevor Hastie and Robert Tibshirani},
    journal = {Annals of statistics},
    number = {2},
    pages = {337-407},
    title = {Additive logistic regression : A statistical view of boosting},
    volume = {28},
    year = {2000},
    ISSN = {0090-5364}
 }

Valid options are:

 -S <num>
  Random number seed.
  (default 1)

 -D
  If set, classifier is run in debug mode and
  may output additional info to the console

 -P <UNPRUNED|POSTPRUNED|PREPRUNED>
  The pruning strategy.
  (default: POSTPRUNED)

 -M <min no>
  The minimal number of instances at the terminal nodes.
  (default 2)

 -N <num folds>
  The number of folds used in the pruning.
  (default 5)

 -H
  Don't use heuristic search for nominal attributes in multi-class
  problem (default yes).

 -G
  Don't use Gini index for splitting (default yes),
  if not information is used.

 -R
  Don't use error rate in internal cross-validation (default yes), 
  but root mean squared error.

 -A
  Use the 1 SE rule to make pruning decision.
  (default no).

 -C
  Percentage of training data size (0-1]
  (default 1).

Version:: $Revision: 6947 $
Author:: Haijian Shi (hs69@cs.waikato.ac.nz)
See Also:: Serialized Form

Field Summary

Fields
Modifier and Type	Field and Description
`protected Attribute`	`m_Attribute` Attribute used for splitting.
`protected Attribute`	`m_ClassAttribute` Class attribute of a dataset.
`protected double[]`	`m_ClassProbs` Class probabilities.
`protected double`	`m_ClassValue` Class value for a node.
`protected double[]`	`m_Distribution` Class distributions.
`protected double[][][]`	`m_Dists` Distributions of each attribute for two successor nodes.
`protected static int`	`m_Expansion` Number of expansions.
`protected int`	`m_FixedExpansion` Fixed number of expansions (if no pruning method is used, its value is -1.
`protected boolean`	`m_Heuristic` If use huristic search for binary split (default true).
`protected boolean`	`m_isLeaf` If the ndoe is leaf node.
`protected int`	`m_minNumObj` Minimum number of instances at leaf nodes.
`protected int`	`m_numFoldsPruning` Number of folds for the pruning.
`protected double[]`	`m_Props` Branch proportions.
`protected int`	`m_PruningStrategy` the pruning strategy
`protected double`	`m_SizePer` The training data size (0-1).
`protected int[][]`	`m_SortedIndices` Sorted indices.
`protected String`	`m_SplitString` Split subset (for nominal attributes).
`protected double`	`m_SplitValue` Split point (for numeric attributes).
`protected BFTree[]`	`m_Successors` Successor nodes.
`protected double`	`m_TotalWeight` Total weights.
`protected boolean`	`m_UseErrorRate` If use error rate in internal cross-validation to fix the number of expansions - default (if not, root mean squared error is used).
`protected boolean`	`m_UseGini` If use Gini index as the splitting criterion - default (if not, information is used).
`protected boolean`	`m_UseOneSE` If use the 1SE rule to make the decision.
`protected double[][]`	`m_Weights` Sorted weights.
`static int`	`PRUNING_POSTPRUNING` pruning strategy: post-pruning
`static int`	`PRUNING_PREPRUNING` pruning strategy: pre-pruning
`static int`	`PRUNING_UNPRUNED` pruning strategy: un-pruned
`static Tag[]`	`TAGS_PRUNING` pruning strategy

Fields inherited from class weka.classifiers.RandomizableClassifier
m_Seed

Fields inherited from class weka.classifiers.Classifier
m_Debug

Constructor Summary

Constructors
Constructor and Description

BFTree()

Constructors
Constructor and Description
`BFTree()`

Method Summary

All Methods Static Methods Instance Methods Concrete Methods
Modifier and Type	Method and Description
`void`	`buildClassifier(Instances data)` Method for building a BestFirst decision tree classifier.
`protected double`	`computeEntropy(double[] dist, double total)` Compute and return entropy for a given distribution of a node.
`protected double`	`computeGini(double[] dist, double total)` Compute and return gini index for a given distribution of a node.
`protected double`	`computeGiniGain(double[] parentDist, double[][] childDist)` Compute and return gini gain for given distributions of a node and its successor nodes.
`protected double`	`computeInfoGain(double[] parentDist, double[][] childDist)` Compute and return information gain for given distributions of a node and its successor nodes.
`protected double`	`computeSortedInfo(Instances data, int[][] sortedIndices, double[][] weights, double[] classProbs)` Compute sorted indices, weights and class probabilities for a given dataset.
`protected FastVector`	`computeSplitInfo(BFTree node, Instances data, int[][] sortedIndices, double[][] weights, double[][][] dists, double[][] props, double[][] totalSubsetWeights, boolean useHeuristic, boolean useGini)` Compute the best splitting attribute, split point or subset and the best gini gain or iformation gain for a given dataset.
`double[]`	`distributionForInstance(Instance instance)` Computes class probabilities for instance using the decision tree.
`Enumeration`	`enumerateMeasures()` Return an enumeration of the measure names.
`Capabilities`	`getCapabilities()` Returns default capabilities of the classifier.
`boolean`	`getHeuristic()` Get if use heuristic search for nominal attributes in multi-class problems.
`double`	`getMeasure(String additionalMeasureName)` Returns the value of the named measure
`int`	`getMinNumObj()` Get minimal number of instances at the terminal nodes.
`int`	`getNumFoldsPruning()` Set number of folds in internal cross-validation.
`String[]`	`getOptions()` Gets the current settings of the Classifier.
`SelectedTag`	`getPruningStrategy()` Gets the pruning strategy.
`String`	`getRevision()` Returns the revision string.
`double`	`getSizePer()` Get training set size.
`TechnicalInformation`	`getTechnicalInformation()` Returns an instance of a TechnicalInformation object, containing detailed information about the technical background of this class, e.g., paper reference or book this class is based on.
`boolean`	`getUseErrorRate()` Get if use error rate in internal cross-validation.
`boolean`	`getUseGini()` Get if use Gini index as splitting criterion.
`boolean`	`getUseOneSE()` Get if use the 1SE rule to choose final model.
`String`	`globalInfo()` Returns a string describing classifier
`String`	`heuristicTipText()` Returns the tip text for this property
`Enumeration`	`listOptions()` Returns an enumeration describing the available options.
`static void`	`main(String[] args)` Main method.
`protected void`	`makeLeaf(Instances data)` Make the node leaf node.
`protected void`	`makeSuccessors(FastVector BestFirstElements, Instances data, int[][][] subsetSortedIndices, double[][][] subsetWeights, double[][][] dists, Attribute att, boolean useHeuristic, boolean useGini)` Generate successor nodes for a node and put them into BestFirstElements according to gini gain or information gain in a descending order.
`protected void`	`makeTree(FastVector BestFirstElements, BFTree root, Instances train, Instances test, FastVector modelError, int[][] sortedIndices, double[][] weights, double[][][] dists, double[] classProbs, double totalWeight, double[] branchProps, int minNumObj, boolean useHeuristic, boolean useGini, boolean useErrorRate)` This method is to find the number of expansions based on internal cross-validation for just post-pruning.
`protected boolean`	`makeTree(FastVector BestFirstElements, BFTree root, Instances train, int[][] sortedIndices, double[][] weights, double[][][] dists, double[] classProbs, double totalWeight, double[] branchProps, int minNumObj, boolean useHeuristic, boolean useGini)` This method is to find the number of expansions based on internal cross-validation for just pre-pruning.
`protected void`	`makeTree(FastVector BestFirstElements, Instances data, int[][] sortedIndices, double[][] weights, double[][][] dists, double[] classProbs, double totalWeight, double[] branchProps, int minNumObj, boolean useHeuristic, boolean useGini, int preExpansion)` Recursively build a best-first decision tree.
`double`	`measureTreeSize()` Return number of tree size.
`String`	`minNumObjTipText()` Returns the tip text for this property
`protected String`	`nominalDistribution(double[][] props, double[][][] dists, Attribute att, int[] sortedIndices, double[] weights, double[][] subsetWeights, double[] gains, Instances data, boolean useHeuristic, boolean useGini)` Compute distributions, proportions and total weights of two successor nodes for a given nominal attribute.
`protected double`	`numericDistribution(double[][] props, double[][][] dists, Attribute att, int[] sortedIndices, double[] weights, double[][] subsetWeights, double[] gains, Instances data, boolean useGini)` Compute distributions, proportions and total weights of two successor nodes for a given numeric attribute.
`String`	`numFoldsPruningTipText()` Returns the tip text for this property
`int`	`numLeaves()` Compute number of leaf nodes.
`int`	`numNodes()` Compute size of the tree.
`String`	`pruningStrategyTipText()` Returns the tip text for this property
`void`	`setHeuristic(boolean value)` Set if use heuristic search for nominal attributes in multi-class problems.
`void`	`setMinNumObj(int value)` Set minimal number of instances at the terminal nodes.
`void`	`setNumFoldsPruning(int value)` Set number of folds in internal cross-validation.
`void`	`setOptions(String[] options)` Parses the options for this object.
`void`	`setPruningStrategy(SelectedTag value)` Sets the pruning strategy.
`void`	`setSizePer(double value)` Set training set size.
`void`	`setUseErrorRate(boolean value)` Set if use error rate in internal cross-validation.
`void`	`setUseGini(boolean value)` Set if use Gini index as splitting criterion.
`void`	`setUseOneSE(boolean value)` Set if use the 1SE rule to choose final model.
`String`	`sizePerTipText()` Returns the tip text for this property
`protected void`	`splitData(int[][][] subsetIndices, double[][][] subsetWeights, Attribute att, double splitPoint, String splitStr, int[][] sortedIndices, double[][] weights, Instances data)` Split data into two subsets and store sorted indices and weights for two successor nodes.
`String`	`toString()` Prints the decision tree using the protected toString method from below.
`protected String`	`toString(int level)` Outputs a tree at a certain level.
`String`	`useErrorRateTipText()` Returns the tip text for this property
`String`	`useGiniTipText()` Returns the tip text for this property
`String`	`useOneSETipText()` Returns the tip text for this property

Methods inherited from class weka.classifiers.RandomizableClassifier
getSeed, seedTipText, setSeed

Methods inherited from class weka.classifiers.Classifier
classifyInstance, debugTipText, forName, getDebug, makeCopies, makeCopy, runClassifier, setDebug

Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

- Field Detail
  - PRUNING_UNPRUNED
```
public static final int PRUNING_UNPRUNED
```
    pruning strategy: un-pruned
    
    See Also:
    
    Constant Field Values
  - PRUNING_POSTPRUNING
```
public static final int PRUNING_POSTPRUNING
```
    pruning strategy: post-pruning
    
    See Also:
    
    Constant Field Values
  - PRUNING_PREPRUNING
```
public static final int PRUNING_PREPRUNING
```
    pruning strategy: pre-pruning
    
    See Also:
    
    Constant Field Values
  - TAGS_PRUNING
```
public static final Tag[] TAGS_PRUNING
```
    pruning strategy
  - m_PruningStrategy
```
protected int m_PruningStrategy
```
    the pruning strategy
  - m_Successors
```
protected BFTree[] m_Successors
```
    Successor nodes.
  - m_Attribute
```
protected Attribute m_Attribute
```
    Attribute used for splitting.
  - m_SplitValue
```
protected double m_SplitValue
```
    Split point (for numeric attributes).
  - m_SplitString
```
protected String m_SplitString
```
    Split subset (for nominal attributes).
  - m_ClassValue
```
protected double m_ClassValue
```
    Class value for a node.
  - m_ClassAttribute
```
protected Attribute m_ClassAttribute
```
    Class attribute of a dataset.
  - m_minNumObj
```
protected int m_minNumObj
```
    Minimum number of instances at leaf nodes.
  - m_numFoldsPruning
```
protected int m_numFoldsPruning
```
    Number of folds for the pruning.
  - m_isLeaf
```
protected boolean m_isLeaf
```
    If the ndoe is leaf node.
  - m_Expansion
```
protected static int m_Expansion
```
    Number of expansions.
  - m_FixedExpansion
```
protected int m_FixedExpansion
```
    Fixed number of expansions (if no pruning method is used, its value is -1. Otherwise, its value is gotten from internal cross-validation).
  - m_Heuristic
```
protected boolean m_Heuristic
```
    If use huristic search for binary split (default true). Note even if its value is true, it is only used when the number of values of a nominal attribute is larger than 4.
  - m_UseGini
```
protected boolean m_UseGini
```
    If use Gini index as the splitting criterion - default (if not, information is used).
  - m_UseErrorRate
```
protected boolean m_UseErrorRate
```
    If use error rate in internal cross-validation to fix the number of expansions - default (if not, root mean squared error is used).
  - m_UseOneSE
```
protected boolean m_UseOneSE
```
    If use the 1SE rule to make the decision.
  - m_Distribution
```
protected double[] m_Distribution
```
    Class distributions.
  - m_Props
```
protected double[] m_Props
```
    Branch proportions.
  - m_SortedIndices
```
protected int[][] m_SortedIndices
```
    Sorted indices.
  - m_Weights
```
protected double[][] m_Weights
```
    Sorted weights.
  - m_Dists
```
protected double[][][] m_Dists
```
    Distributions of each attribute for two successor nodes.
  - m_ClassProbs
```
protected double[] m_ClassProbs
```
    Class probabilities.
  - m_TotalWeight
```
protected double m_TotalWeight
```
    Total weights.
  - m_SizePer
```
protected double m_SizePer
```
    The training data size (0-1). Default 1.
- Constructor Detail
  - BFTree
```
public BFTree()
```
- Method Detail
  - globalInfo
```
public String globalInfo()
```
    Returns a string describing classifier
    
    Returns:
    
    a description suitable for displaying in the explorer/experimenter gui
  - getTechnicalInformation
```
public TechnicalInformation getTechnicalInformation()
```
    Returns an instance of a TechnicalInformation object, containing detailed information about the technical background of this class, e.g., paper reference or book this class is based on.
    
    Specified by:
    
    getTechnicalInformation in interface TechnicalInformationHandler
    
    Returns:
    
    the technical information about this class
  - getCapabilities
```
public Capabilities getCapabilities()
```
    Returns default capabilities of the classifier.
    
    Specified by:
    
    getCapabilities in interface CapabilitiesHandler
    
    Overrides:
    
    getCapabilities in class Classifier
    
    Returns:
    
    the capabilities of this classifier
    
    See Also:
    
    Capabilities
  - buildClassifier
```
public void buildClassifier(Instances data)
                     throws Exception
```
    Method for building a BestFirst decision tree classifier.
    
    Specified by:
    
    buildClassifier in class Classifier
    
    Parameters:
    
    data - set of instances serving as training data
    
    Throws:
    
    Exception - if decision tree cannot be built successfully
  - makeTree
```
protected void makeTree(FastVector BestFirstElements,
                        Instances data,
                        int[][] sortedIndices,
                        double[][] weights,
                        double[][][] dists,
                        double[] classProbs,
                        double totalWeight,
                        double[] branchProps,
                        int minNumObj,
                        boolean useHeuristic,
                        boolean useGini,
                        int preExpansion)
                 throws Exception
```
    Recursively build a best-first decision tree. Method for building a Best-First tree for a given number of expansions. preExpasion is -1 means that no expansion is specified (just for a tree without any pruning method). Pre-pruning and post-pruning methods also use this method to build the final tree on all training data based on the expansion calculated from internal cross-validation.
    
    Parameters:
    
    BestFirstElements - list to store BFTree nodes
    
    data - training data
    
    sortedIndices - sorted indices of the instances
    
    weights - weights of the instances
    
    dists - class distributions for each attribute
    
    classProbs - class probabilities of this node
    
    totalWeight - total weight of this node (note if the node can not split, this value is not calculated.)
    
    branchProps - proportions of two subbranches
    
    minNumObj - minimal number of instances at leaf nodes
    
    useHeuristic - if use heuristic search for nominal attributes in multi-class problem
    
    useGini - if use Gini index as splitting criterion
    
    preExpansion - the number of expansions the tree to be expanded
    
    Throws:
    
    Exception - if something goes wrong
  - makeTree
```
protected boolean makeTree(FastVector BestFirstElements,
                           BFTree root,
                           Instances train,
                           int[][] sortedIndices,
                           double[][] weights,
                           double[][][] dists,
                           double[] classProbs,
                           double totalWeight,
                           double[] branchProps,
                           int minNumObj,
                           boolean useHeuristic,
                           boolean useGini)
                    throws Exception
```
    This method is to find the number of expansions based on internal cross-validation for just pre-pruning. It expands the first BestFirst node in the BestFirstElements if it is expansible, otherwise it looks for next exapansible node. If it finds a node is expansibel, expand the node, then return true. (note it just expands one node at a time).
    
    Parameters:
    
    BestFirstElements - list to store BFTree nodes
    
    root - root node of tree in each fold
    
    train - training data
    
    sortedIndices - sorted indices of the instances
    
    weights - weights of the instances
    
    dists - class distributions for each attribute
    
    classProbs - class probabilities of this node
    
    totalWeight - total weight of this node (note if the node can not split, this value is not calculated.)
    
    branchProps - proportions of two subbranches
    
    minNumObj - minimal number of instances at leaf nodes
    
    useHeuristic - if use heuristic search for nominal attributes in multi-class problem
    
    useGini - if use Gini index as splitting criterion
    
    Returns:
    
    true if expand successfully, otherwise return false (all nodes in BestFirstElements cannot be expanded).
    
    Throws:
    
    Exception - if something goes wrong
  - makeTree
```
protected void makeTree(FastVector BestFirstElements,
                        BFTree root,
                        Instances train,
                        Instances test,
                        FastVector modelError,
                        int[][] sortedIndices,
                        double[][] weights,
                        double[][][] dists,
                        double[] classProbs,
                        double totalWeight,
                        double[] branchProps,
                        int minNumObj,
                        boolean useHeuristic,
                        boolean useGini,
                        boolean useErrorRate)
                 throws Exception
```
    This method is to find the number of expansions based on internal cross-validation for just post-pruning. It expands the first BestFirst node in the BestFirstElements until no node can be split. When building the tree, stroe error for each temporary tree, namely for each expansion.
    
    Parameters:
    
    BestFirstElements - list to store BFTree nodes
    
    root - root node of tree in each fold
    
    train - training data in each fold
    
    test - test data in each fold
    
    modelError - list to store error for each expansion in each fold
    
    sortedIndices - sorted indices of the instances
    
    weights - weights of the instances
    
    dists - class distributions for each attribute
    
    classProbs - class probabilities of this node
    
    totalWeight - total weight of this node (note if the node can not split, this value is not calculated.)
    
    branchProps - proportions of two subbranches
    
    minNumObj - minimal number of instances at leaf nodes
    
    useHeuristic - if use heuristic search for nominal attributes in multi-class problem
    
    useGini - if use Gini index as splitting criterion
    
    useErrorRate - if use error rate in internal cross-validation
    
    Throws:
    
    Exception - if something goes wrong
  - makeSuccessors
```
protected void makeSuccessors(FastVector BestFirstElements,
                              Instances data,
                              int[][][] subsetSortedIndices,
                              double[][][] subsetWeights,
                              double[][][] dists,
                              Attribute att,
                              boolean useHeuristic,
                              boolean useGini)
                       throws Exception
```
    Generate successor nodes for a node and put them into BestFirstElements according to gini gain or information gain in a descending order.
    
    Parameters:
    
    BestFirstElements - list to store BestFirst nodes
    
    data - training instance
    
    subsetSortedIndices - sorted indices of instances of successor nodes
    
    subsetWeights - weights of instances of successor nodes
    
    dists - class distributions of successor nodes
    
    att - attribute used to split the node
    
    useHeuristic - if use heuristic search for nominal attributes in multi-class problem
    
    useGini - if use Gini index as splitting criterion
    
    Throws:
    
    Exception - if something goes wrong
  - computeSortedInfo
```
protected double computeSortedInfo(Instances data,
                                   int[][] sortedIndices,
                                   double[][] weights,
                                   double[] classProbs)
                            throws Exception
```
    Compute sorted indices, weights and class probabilities for a given dataset. Return total weights of the data at the node.
    
    Parameters:
    
    data - training data
    
    sortedIndices - sorted indices of instances at the node
    
    weights - weights of instances at the node
    
    classProbs - class probabilities at the node
    
    Returns:
    
    total weights of instances at the node
    
    Throws:
    
    Exception - if something goes wrong
  - computeSplitInfo
```
protected FastVector computeSplitInfo(BFTree node,
                                      Instances data,
                                      int[][] sortedIndices,
                                      double[][] weights,
                                      double[][][] dists,
                                      double[][] props,
                                      double[][] totalSubsetWeights,
                                      boolean useHeuristic,
                                      boolean useGini)
                               throws Exception
```
    Compute the best splitting attribute, split point or subset and the best gini gain or iformation gain for a given dataset.
    
    Parameters:
    
    node - node to be split
    
    data - training data
    
    sortedIndices - sorted indices of the instances
    
    weights - weights of the instances
    
    dists - class distributions for each attribute
    
    props - proportions of two branches
    
    totalSubsetWeights - total weight of two subsets
    
    useHeuristic - if use heuristic search for nominal attributes in multi-class problem
    
    useGini - if use Gini index as splitting criterion
    
    Returns:
    
    split information about the node
    
    Throws:
    
    Exception - if something is wrong
  - numericDistribution
```
protected double numericDistribution(double[][] props,
                                     double[][][] dists,
                                     Attribute att,
                                     int[] sortedIndices,
                                     double[] weights,
                                     double[][] subsetWeights,
                                     double[] gains,
                                     Instances data,
                                     boolean useGini)
                              throws Exception
```
    Compute distributions, proportions and total weights of two successor nodes for a given numeric attribute.
    
    Parameters:
    
    props - proportions of each two branches for each attribute
    
    dists - class distributions of two branches for each attribute
    
    att - numeric att split on
    
    sortedIndices - sorted indices of instances for the attirubte
    
    weights - weights of instances for the attirbute
    
    subsetWeights - total weight of two branches split based on the attribute
    
    gains - Gini gains or information gains for each attribute
    
    data - training instances
    
    useGini - if use Gini index as splitting criterion
    
    Returns:
    
    Gini gain or information gain for the given attribute
    
    Throws:
    
    Exception - if something goes wrong
  - nominalDistribution
```
protected String nominalDistribution(double[][] props,
                                     double[][][] dists,
                                     Attribute att,
                                     int[] sortedIndices,
                                     double[] weights,
                                     double[][] subsetWeights,
                                     double[] gains,
                                     Instances data,
                                     boolean useHeuristic,
                                     boolean useGini)
                              throws Exception
```
    Compute distributions, proportions and total weights of two successor nodes for a given nominal attribute.
    
    Parameters:
    
    props - proportions of each two branches for each attribute
    
    dists - class distributions of two branches for each attribute
    
    att - numeric att split on
    
    sortedIndices - sorted indices of instances for the attirubte
    
    weights - weights of instances for the attirbute
    
    subsetWeights - total weight of two branches split based on the attribute
    
    gains - Gini gains for each attribute
    
    data - training instances
    
    useHeuristic - if use heuristic search
    
    useGini - if use Gini index as splitting criterion
    
    Returns:
    
    Gini gain for the given attribute
    
    Throws:
    
    Exception - if something goes wrong
  - splitData
```
protected void splitData(int[][][] subsetIndices,
                         double[][][] subsetWeights,
                         Attribute att,
                         double splitPoint,
                         String splitStr,
                         int[][] sortedIndices,
                         double[][] weights,
                         Instances data)
                  throws Exception
```
    Split data into two subsets and store sorted indices and weights for two successor nodes.
    
    Parameters:
    
    subsetIndices - sorted indecis of instances for each attribute for two successor node
    
    subsetWeights - weights of instances for each attribute for two successor node
    
    att - attribute the split based on
    
    splitPoint - split point the split based on if att is numeric
    
    splitStr - split subset the split based on if att is nominal
    
    sortedIndices - sorted indices of the instances to be split
    
    weights - weights of the instances to bes split
    
    data - training data
    
    Throws:
    
    Exception - if something goes wrong
  - computeGiniGain
```
protected double computeGiniGain(double[] parentDist,
                                 double[][] childDist)
```
    Compute and return gini gain for given distributions of a node and its successor nodes.
    
    Parameters:
    
    parentDist - class distributions of parent node
    
    childDist - class distributions of successor nodes
    
    Returns:
    
    Gini gain computed
  - computeGini
```
protected double computeGini(double[] dist,
                             double total)
```
    Compute and return gini index for a given distribution of a node.
    
    Parameters:
    
    dist - class distributions
    
    total - class distributions
    
    Returns:
    
    Gini index of the class distributions
  - computeInfoGain
```
protected double computeInfoGain(double[] parentDist,
                                 double[][] childDist)
```
    Compute and return information gain for given distributions of a node and its successor nodes.
    
    Parameters:
    
    parentDist - class distributions of parent node
    
    childDist - class distributions of successor nodes
    
    Returns:
    
    information gain computed
  - computeEntropy
```
protected double computeEntropy(double[] dist,
                                double total)
```
    Compute and return entropy for a given distribution of a node.
    
    Parameters:
    
    dist - class distributions
    
    total - class distributions
    
    Returns:
    
    entropy of the class distributions
  - makeLeaf
```
protected void makeLeaf(Instances data)
```
    Make the node leaf node.
    
    Parameters:
    
    data - training data
  - distributionForInstance
```
public double[] distributionForInstance(Instance instance)
                                 throws Exception
```
    Computes class probabilities for instance using the decision tree.
    
    Overrides:
    
    distributionForInstance in class Classifier
    
    Parameters:
    
    instance - the instance for which class probabilities is to be computed
    
    Returns:
    
    the class probabilities for the given instance
    
    Throws:
    
    Exception - if something goes wrong
  - toString
```
public String toString()
```
    Prints the decision tree using the protected toString method from below.
    
    Overrides:
    
    toString in class Object
    
    Returns:
    
    a textual description of the classifier
  - toString
```
protected String toString(int level)
```
    Outputs a tree at a certain level.
    
    Parameters:
    
    level - the level at which the tree is to be printed
    
    Returns:
    
    a tree at a certain level.
  - numNodes
```
public int numNodes()
```
    Compute size of the tree.
    
    Returns:
    
    size of the tree
  - numLeaves
```
public int numLeaves()
```
    Compute number of leaf nodes.
    
    Returns:
    
    number of leaf nodes
  - listOptions
```
public Enumeration listOptions()
```
    Returns an enumeration describing the available options.
    
    Specified by:
    
    listOptions in interface OptionHandler
    
    Overrides:
    
    listOptions in class RandomizableClassifier
    
    Returns:
    
    an enumeration describing the available options.
  - setOptions
```
public void setOptions(String[] options)
                throws Exception
```
    Parses the options for this object.
    Valid options are:
```
 -S <num>
  Random number seed.
  (default 1)
```
```
 -D
  If set, classifier is run in debug mode and
  may output additional info to the console
```
```
 -P <UNPRUNED|POSTPRUNED|PREPRUNED>
  The pruning strategy.
  (default: POSTPRUNED)
```
```
 -M <min no>
  The minimal number of instances at the terminal nodes.
  (default 2)
```
```
 -N <num folds>
  The number of folds used in the pruning.
  (default 5)
```
```
 -H
  Don't use heuristic search for nominal attributes in multi-class
  problem (default yes).
 
```
```
 -G
  Don't use Gini index for splitting (default yes),
  if not information is used.
```
```
 -R
  Don't use error rate in internal cross-validation (default yes), 
  but root mean squared error.
```
```
 -A
  Use the 1 SE rule to make pruning decision.
  (default no).
```
```
 -C
  Percentage of training data size (0-1]
  (default 1).
```
    Specified by:
    
    setOptions in interface OptionHandler
    
    Overrides:
    
    setOptions in class RandomizableClassifier
    
    Parameters:
    
    options - the options to use
    
    Throws:
    
    Exception - if setting of options fails
  - getOptions
```
public String[] getOptions()
```
    Gets the current settings of the Classifier.
    
    Specified by:
    
    getOptions in interface OptionHandler
    
    Overrides:
    
    getOptions in class RandomizableClassifier
    
    Returns:
    
    the current settings of the Classifier
  - enumerateMeasures
```
public Enumeration enumerateMeasures()
```
    Return an enumeration of the measure names.
    
    Specified by:
    
    enumerateMeasures in interface AdditionalMeasureProducer
    
    Returns:
    
    an enumeration of the measure names
  - measureTreeSize
```
public double measureTreeSize()
```
    Return number of tree size.
    
    Returns:
    
    number of tree size
  - getMeasure
```
public double getMeasure(String additionalMeasureName)
```
    Returns the value of the named measure
    
    Specified by:
    
    getMeasure in interface AdditionalMeasureProducer
    
    Parameters:
    
    additionalMeasureName - the name of the measure to query for its value
    
    Returns:
    
    the value of the named measure
    
    Throws:
    
    IllegalArgumentException - if the named measure is not supported
  - pruningStrategyTipText
```
public String pruningStrategyTipText()
```
    Returns the tip text for this property
    
    Returns:
    
    tip text for this property suitable for displaying in the explorer/experimenter gui
  - setPruningStrategy
```
public void setPruningStrategy(SelectedTag value)
```
    Sets the pruning strategy.
    
    Parameters:
    
    value - the strategy
  - getPruningStrategy
```
public SelectedTag getPruningStrategy()
```
    Gets the pruning strategy.
    
    Returns:
    
    the current strategy.
  - minNumObjTipText
```
public String minNumObjTipText()
```
    Returns the tip text for this property
    
    Returns:
    
    tip text for this property suitable for displaying in the explorer/experimenter gui
  - setMinNumObj
```
public void setMinNumObj(int value)
```
    Set minimal number of instances at the terminal nodes.
    
    Parameters:
    
    value - minimal number of instances at the terminal nodes
  - getMinNumObj
```
public int getMinNumObj()
```
    Get minimal number of instances at the terminal nodes.
    
    Returns:
    
    minimal number of instances at the terminal nodes
  - numFoldsPruningTipText
```
public String numFoldsPruningTipText()
```
    Returns the tip text for this property
    
    Returns:
    
    tip text for this property suitable for displaying in the explorer/experimenter gui
  - setNumFoldsPruning
```
public void setNumFoldsPruning(int value)
```
    Set number of folds in internal cross-validation.
    
    Parameters:
    
    value - the number of folds
  - getNumFoldsPruning
```
public int getNumFoldsPruning()
```
    Set number of folds in internal cross-validation.
    
    Returns:
    
    number of folds in internal cross-validation
  - heuristicTipText
```
public String heuristicTipText()
```
    Returns the tip text for this property
    
    Returns:
    
    tip text for this property suitable for displaying in the explorer/experimenter gui.
  - setHeuristic
```
public void setHeuristic(boolean value)
```
    Set if use heuristic search for nominal attributes in multi-class problems.
    
    Parameters:
    
    value - if use heuristic search for nominal attributes in multi-class problems
  - getHeuristic
```
public boolean getHeuristic()
```
    Get if use heuristic search for nominal attributes in multi-class problems.
    
    Returns:
    
    if use heuristic search for nominal attributes in multi-class problems
  - useGiniTipText
```
public String useGiniTipText()
```
    Returns the tip text for this property
    
    Returns:
    
    tip text for this property suitable for displaying in the explorer/experimenter gui.
  - setUseGini
```
public void setUseGini(boolean value)
```
    Set if use Gini index as splitting criterion.
    
    Parameters:
    
    value - if use Gini index splitting criterion
  - getUseGini
```
public boolean getUseGini()
```
    Get if use Gini index as splitting criterion.
    
    Returns:
    
    if use Gini index as splitting criterion
  - useErrorRateTipText
```
public String useErrorRateTipText()
```
    Returns the tip text for this property
    
    Returns:
    
    tip text for this property suitable for displaying in the explorer/experimenter gui.
  - setUseErrorRate
```
public void setUseErrorRate(boolean value)
```
    Set if use error rate in internal cross-validation.
    
    Parameters:
    
    value - if use error rate in internal cross-validation
  - getUseErrorRate
```
public boolean getUseErrorRate()
```
    Get if use error rate in internal cross-validation.
    
    Returns:
    
    if use error rate in internal cross-validation.
  - useOneSETipText
```
public String useOneSETipText()
```
    Returns the tip text for this property
    
    Returns:
    
    tip text for this property suitable for displaying in the explorer/experimenter gui.
  - setUseOneSE
```
public void setUseOneSE(boolean value)
```
    Set if use the 1SE rule to choose final model.
    
    Parameters:
    
    value - if use the 1SE rule to choose final model
  - getUseOneSE
```
public boolean getUseOneSE()
```
    Get if use the 1SE rule to choose final model.
    
    Returns:
    
    if use the 1SE rule to choose final model
  - sizePerTipText
```
public String sizePerTipText()
```
    Returns the tip text for this property
    
    Returns:
    
    tip text for this property suitable for displaying in the explorer/experimenter gui.
  - setSizePer
```
public void setSizePer(double value)
```
    Set training set size.
    
    Parameters:
    
    value - training set size
  - getSizePer
```
public double getSizePer()
```
    Get training set size.
    
    Returns:
    
    training set size
  - getRevision
```
public String getRevision()
```
    Returns the revision string.
    
    Specified by:
    
    getRevision in interface RevisionHandler
    
    Overrides:
    
    getRevision in class Classifier
    
    Returns:
    
    the revision
  - main
```
public static void main(String[] args)
```
    Main method.
    
    Parameters:
    
    args - the options for the classifier

Class BFTree

Field Summary

Fields inherited from class weka.classifiers.RandomizableClassifier

Fields inherited from class weka.classifiers.Classifier

Constructor Summary

Method Summary

Methods inherited from class weka.classifiers.RandomizableClassifier

Methods inherited from class weka.classifiers.Classifier

Methods inherited from class java.lang.Object

Field Detail

PRUNING_UNPRUNED

PRUNING_POSTPRUNING

PRUNING_PREPRUNING

TAGS_PRUNING

m_PruningStrategy

m_Successors

m_Attribute

m_SplitValue

m_SplitString

m_ClassValue

m_ClassAttribute

m_minNumObj

m_numFoldsPruning

m_isLeaf

m_Expansion

m_FixedExpansion

m_Heuristic

m_UseGini

m_UseErrorRate

m_UseOneSE

m_Distribution

m_Props

m_SortedIndices

m_Weights

m_Dists

m_ClassProbs

m_TotalWeight

m_SizePer

Constructor Detail

BFTree

Method Detail

globalInfo

getTechnicalInformation

getCapabilities

buildClassifier

makeTree

makeTree

makeTree

makeSuccessors

computeSortedInfo

computeSplitInfo

numericDistribution

nominalDistribution

splitData

computeGiniGain

computeGini

computeInfoGain

computeEntropy

makeLeaf

distributionForInstance

toString

toString

numNodes

numLeaves

listOptions

setOptions

getOptions

enumerateMeasures

measureTreeSize

getMeasure

pruningStrategyTipText

setPruningStrategy

getPruningStrategy

minNumObjTipText

setMinNumObj

getMinNumObj

numFoldsPruningTipText

setNumFoldsPruning

getNumFoldsPruning

heuristicTipText