weka.classifiers
Class MetaCost

java.lang.Object
  |
  +--weka.classifiers.Classifier
        |
        +--weka.classifiers.MetaCost

All Implemented Interfaces:

java.lang.Cloneable, OptionHandler, java.io.Serializable

public class MetaCost

extends Classifier

implements OptionHandler

This metaclassifier makes its base classifier cost-sensitive using the method specified in

Pedro Domingos (1999). MetaCost: A general method for making classifiers cost-sensitive, Proceedings of the Fifth International Conference on Knowledge Discovery and Data Mining, pp. 155-164. Also available online at http://www.cs.washington.edu/homes/pedrod/kdd99.ps.gz.

This classifier should produce similar results to one created by passing the base learner to Bagging, which is in turn passed to a CostSensitiveClassifier operating on minimum expected cost. The difference is that MetaCost produces a single cost-sensitive classifier of the base learner, giving the benefits of fast classification and interpretable output (if the base learner itself is interpretable). This implementation uses all bagging iterations when reclassifying training data (the MetaCost paper reports a marginal improvement when only those iterations containing each training instance are used in reclassifying that instance).

Valid options are:

-W classname
Specify the full class name of a classifier (required).

-C cost file
File name of a cost matrix to use. If this is not supplied, a cost matrix will be loaded on demand. The name of the on-demand file is the relation name of the training data plus ".cost", and the path to the on-demand file is specified with the -D option.

-D directory
Name of a directory to search for cost files when loading costs on demand (default current directory).

-I num
Set the number of bagging iterations (default 10).

-S seed
Random number seed used when reweighting by resampling (default 1).

-P num
Size of each bag, as a percentage of the training size (default 100).

Options after -- are passed to the designated classifier.

Author:

Len Trigg (len@intelligenesis.net)

See Also:

Serialized Form

Field Summary

protected int	m_BagSizePercent The size of each bag sample, as a percentage of the training size
protected Classifier	m_Classifier The classifier
protected java.lang.String	m_CostFile The name of the cost file, for command line options
protected CostMatrix	m_CostMatrix The cost matrix
protected int	m_MatrixSource Indicates the current cost matrix source
protected int	m_NumIterations The number of iterations.
protected java.io.File	m_OnDemandDirectory The directory used when loading cost files on demand, null indicates current directory
protected int	m_Seed Seed for reweighting using resampling.
static int	MATRIX_ON_DEMAND
static int	MATRIX_SUPPLIED
static Tag[]	TAGS_MATRIX_SOURCE

Constructor Summary

MetaCost()

Method Summary

void	buildClassifier(Instances data) Builds the model of the base learner.
double	classifyInstance(Instance instance) Classifies a given test instance.
int	getBagSizePercent() Gets the size of each bag, as a percentage of the training set size.
Classifier	getClassifier() Gets the distribution classifier used.
protected java.lang.String	getClassifierSpec() Gets the classifier specification string, which contains the class name of the classifier and any options to the classifier
CostMatrix	getCostMatrix() Gets the misclassification cost matrix.
SelectedTag	getCostMatrixSource() Gets the source location method of the cost matrix.
int	getNumIterations() Gets the number of bagging iterations
java.io.File	getOnDemandDirectory() Returns the directory that will be searched for cost files when loading on demand.
java.lang.String[]	getOptions() Gets the current settings of the Classifier.
int	getSeed() Get seed for resampling.
java.util.Enumeration	listOptions() Returns an enumeration describing the available options
static void	main(java.lang.String[] argv) Main method for testing this class.
void	setBagSizePercent(int newBagSizePercent) Sets the size of each bag, as a percentage of the training set size.
void	setClassifier(Classifier classifier) Sets the distribution classifier
void	setCostMatrix(CostMatrix newCostMatrix) Sets the misclassification cost matrix.
void	setCostMatrixSource(SelectedTag newMethod) Sets the source location of the cost matrix.
void	setNumIterations(int numIterations) Sets the number of bagging iterations
void	setOnDemandDirectory(java.io.File newDir) Sets the directory that will be searched for cost files when loading on demand.
void	setOptions(java.lang.String[] options) Parses a given list of options.
void	setSeed(int seed) Set seed for resampling.
java.lang.String	toString() Output a representation of this classifier

Methods inherited from class weka.classifiers.Classifier

forName, makeCopies

Methods inherited from class java.lang.Object

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

Field Detail

MATRIX_ON_DEMAND

public static final int MATRIX_ON_DEMAND

MATRIX_SUPPLIED

public static final int MATRIX_SUPPLIED

TAGS_MATRIX_SOURCE

public static final Tag[] TAGS_MATRIX_SOURCE

m_MatrixSource

protected int m_MatrixSource

Indicates the current cost matrix source

m_OnDemandDirectory

protected java.io.File m_OnDemandDirectory

The directory used when loading cost files on demand, null indicates current directory

m_CostFile

protected java.lang.String m_CostFile

The name of the cost file, for command line options

m_Classifier

protected Classifier m_Classifier

The classifier

m_CostMatrix

protected CostMatrix m_CostMatrix

The cost matrix

m_NumIterations

protected int m_NumIterations

The number of iterations.

m_Seed

protected int m_Seed

Seed for reweighting using resampling.

m_BagSizePercent

protected int m_BagSizePercent

The size of each bag sample, as a percentage of the training size

Constructor Detail

MetaCost

public MetaCost()

Method Detail

listOptions

public java.util.Enumeration listOptions()

Returns an enumeration describing the available options

Specified by:

listOptions in interface OptionHandler

Returns:

an enumeration of all the available options

setOptions

public void setOptions(java.lang.String[] options)
                throws java.lang.Exception

Parses a given list of options. Valid options are:

-W classname
Specify the full class name of a classifier (required).

-C cost file
File name of a cost matrix to use. If this is not supplied, a cost matrix will be loaded on demand. The name of the on-demand file is the relation name of the training data plus ".cost", and the path to the on-demand file is specified with the -D option.

-D directory
Name of a directory to search for cost files when loading costs on demand (default current directory).

-I num
Set the number of bagging iterations (default 10).

-S seed
Random number seed used when reweighting by resampling (default 1).

-P num
Size of each bag, as a percentage of the training size (default 100).

Options after -- are passed to the designated classifier.

Specified by:

setOptions in interface OptionHandler

Parameters:

options - the list of options as an array of strings

Throws:

java.lang.Exception - if an option is not supported

getOptions

public java.lang.String[] getOptions()

Gets the current settings of the Classifier.

Specified by:

getOptions in interface OptionHandler

Returns:

an array of strings suitable for passing to setOptions

getCostMatrixSource

public SelectedTag getCostMatrixSource()

Gets the source location method of the cost matrix. Will be one of MATRIX_ON_DEMAND or MATRIX_SUPPLIED.

Returns:

the cost matrix source.

setCostMatrixSource

public void setCostMatrixSource(SelectedTag newMethod)

Sets the source location of the cost matrix. Values other than MATRIX_ON_DEMAND or MATRIX_SUPPLIED will be ignored.

Parameters:

newMethod - the cost matrix location method.

getOnDemandDirectory

public java.io.File getOnDemandDirectory()

Returns the directory that will be searched for cost files when loading on demand.

Returns:

The cost file search directory.

setOnDemandDirectory

public void setOnDemandDirectory(java.io.File newDir)

Sets the directory that will be searched for cost files when loading on demand.

Parameters:

newDir - The cost file search directory.

setClassifier

public void setClassifier(Classifier classifier)

Sets the distribution classifier

Parameters:

classifier - the distribution classifier with all options set.

getClassifier

public Classifier getClassifier()

Gets the distribution classifier used.

Returns:

the classifier

getClassifierSpec

protected java.lang.String getClassifierSpec()

Gets the classifier specification string, which contains the class name of the classifier and any options to the classifier

Returns:

the classifier string.

getBagSizePercent

public int getBagSizePercent()

Gets the size of each bag, as a percentage of the training set size.

Returns:

the bag size, as a percentage.

setBagSizePercent

public void setBagSizePercent(int newBagSizePercent)

Sets the size of each bag, as a percentage of the training set size.

Parameters:

newBagSizePercent - the bag size, as a percentage.

setNumIterations

public void setNumIterations(int numIterations)

Sets the number of bagging iterations

getNumIterations

public int getNumIterations()

Gets the number of bagging iterations

Returns:

the maximum number of bagging iterations

getCostMatrix

public CostMatrix getCostMatrix()

Gets the misclassification cost matrix.

Returns:

the cost matrix

setCostMatrix

public void setCostMatrix(CostMatrix newCostMatrix)

Sets the misclassification cost matrix.

Parameters:

the - cost matrix

setSeed

public void setSeed(int seed)

Set seed for resampling.

Parameters:

seed - the seed for resampling

getSeed

public int getSeed()

Get seed for resampling.

Returns:

the seed for resampling

buildClassifier

public void buildClassifier(Instances data)
                     throws java.lang.Exception

Builds the model of the base learner.

Overrides:

buildClassifier in class Classifier

Parameters:

data - the training data

Throws:

java.lang.Exception - if the classifier could not be built successfully

classifyInstance

public double classifyInstance(Instance instance)
                        throws java.lang.Exception

Classifies a given test instance.

Overrides:

classifyInstance in class Classifier

Parameters:

instance - the instance to be classified

Throws:

java.lang.Exception - if instance could not be classified successfully

toString

public java.lang.String toString()

Output a representation of this classifier

Overrides:

toString in class java.lang.Object

main

public static void main(java.lang.String[] argv)

Main method for testing this class.

Parameters:

argv - should contain the following arguments: -t training file [-T test file] [-c class index]