weka.classifiers.trees.ft

Class FTNode

java.lang.Object

weka.classifiers.Classifier

weka.classifiers.trees.lmt.LogisticBase

weka.classifiers.trees.ft.FTtree

weka.classifiers.trees.ft.FTNode

All Implemented Interfaces:

Serializable, Cloneable, CapabilitiesHandler, OptionHandler, RevisionHandler, WeightedInstancesHandler

public class FTNode
extends FTtree

Class for Functional tree structure.

Version:

$Revision: 1.4 $

Author:

Jo\~{a}o Gama, Carlos Ferreira

See Also:

Serialized Form

Field Summary

Fields inherited from class weka.classifiers.trees.ft.FTtree

m_auxLocalModel, m_CF, m_constError, m_hasConstr, m_higherRegressions, m_id, m_isLeaf, m_leafclass, m_leafModelNum, m_localModel, m_minNumInstances, m_modelSelection, m_nominalToBinary, m_numHigherRegressions, m_numInstances, m_sons, m_totalInstanceWeight

Fields inherited from class weka.classifiers.trees.lmt.LogisticBase

m_errorOnProbabilities, m_fixedNumIterations, m_heuristicStop, m_maxIterations, m_numClasses, m_numericData, m_numericDataHeader, m_numFoldsBoosting, m_numParameters, m_numRegressions, m_regressions, m_train, m_useCrossValidation, m_weightTrimBeta, Z_MAX

Fields inherited from class weka.classifiers.Classifier

m_Debug

Constructor Summary

Constructors

Constructor and Description
FTNode(boolean errorOnProbabilities, int numBoostingIterations, int minNumInstances, double weightTrimBeta, boolean useAIC) Constructor for Functional tree node.

Method Summary

Modifier and Type	Method and Description
void	buildClassifier(Instances data) Method for building a Functional tree (only called for the root node).
void	buildTree(Instances data, SimpleLinearRegression[][] higherRegressions, double totalInstanceWeight, double higherNumParameters) Method for building the tree structure.
double[]	distributionForInstance(Instance instance) Returns the class probabilities for an instance given by the Functional Tree.
String	getRevision() Returns the revision string.
double	prune() Method for prunning a tree using C4.5 pruning procedure.

Methods inherited from class weka.classifiers.trees.ft.FTtree

assignIDs, assignLeafModelNumbers, cleanup, dumpTree, getCoefficients, getConstError, getEstimatedErrors, getEstimatedErrorsForBranch, getEstimatedErrorsForDistribution, getEtimateConstModel, getFs, getModelParameters, getNodes, getNodes, getNumericData, getNumInnerNodes, getNumLeaves, graph, graphTree, hasModels, insertNewAttr, mergeArrays, modelDistributionForInstance, modelsToString, numLeaves, numNodes, removeExtAttributes, toString

Methods inherited from class weka.classifiers.trees.lmt.LogisticBase

getBestIteration, getErrorRate, getFs, getMaxIterations, getMeanAbsoluteError, getNumRegressions, getProbs, getUseAIC, getUsedAttributes, getWeightTrimBeta, getWs, getYs, getZ, getZs, initRegressions, negativeLogLikelihood, percentAttributesUsed, performBoosting, performBoosting, performBoosting, performBoostingCV, performBoostingInfCriterion, performIteration, probs, selectRegressions, setHeuristicStop, setMaxIterations, setUseAIC, setWeightTrimBeta

Methods inherited from class weka.classifiers.Classifier

classifyInstance, debugTipText, forName, getCapabilities, getDebug, getOptions, listOptions, makeCopies, makeCopy, runClassifier, setDebug, setOptions

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

Constructor Detail

FTNode

public FTNode(boolean errorOnProbabilities,
              int numBoostingIterations,
              int minNumInstances,
              double weightTrimBeta,
              boolean useAIC)

Constructor for Functional tree node.

Parameters:

errorOnProbabilities - Use error on probabilities for stopping criterion of LogitBoost?

numBoostingIterations - sets the numBoostingIterations parameter

minNumInstances - minimum number of instances at which a node is considered for splitting

Method Detail

buildClassifier

public void buildClassifier(Instances data)
                     throws Exception

Method for building a Functional tree (only called for the root node). Grows an initial Functional Tree.

Specified by:

buildClassifier in class FTtree

Parameters:

data - the data to train with

Throws:

Exception - if something goes wrong

buildTree

public void buildTree(Instances data,
                      SimpleLinearRegression[][] higherRegressions,
                      double totalInstanceWeight,
                      double higherNumParameters)
               throws Exception

Method for building the tree structure. Builds a logistic model, splits the node and recursively builds tree for child nodes.

Specified by:

buildTree in class FTtree

Parameters:

data - the training data passed on to this node

higherRegressions - An array of regression functions produced by LogitBoost at higher levels in the tree. They represent a logistic regression model that is refined locally at this node.

totalInstanceWeight - the total number of training examples

higherNumParameters - effective number of parameters in the logistic regression model built in parent nodes

Throws:

Exception - if something goes wrong

prune

public double prune()
             throws Exception

Method for prunning a tree using C4.5 pruning procedure.

Specified by:

prune in class FTtree

Throws:

Exception - if something goes wrong

distributionForInstance

public double[] distributionForInstance(Instance instance)
                                 throws Exception

Returns the class probabilities for an instance given by the Functional Tree.

Specified by:

distributionForInstance in class FTtree

Parameters:

instance - the instance

Returns:

the array of probabilities

Throws:

Exception - if distribution can't be computed successfully

getRevision

public String getRevision()

Returns the revision string.

Specified by:

getRevision in interface RevisionHandler

Overrides:

getRevision in class FTtree

Returns:

the revision