weka.classifiers.bayes.net.search.local

Class LocalScoreSearchAlgorithm

java.lang.Object

weka.classifiers.bayes.net.search.SearchAlgorithm

weka.classifiers.bayes.net.search.local.LocalScoreSearchAlgorithm

All Implemented Interfaces:

java.io.Serializable, OptionHandler, RevisionHandler

Direct Known Subclasses:

CISearchAlgorithm, GeneticSearch, HillClimber, K2, SimulatedAnnealing, TAN

public class LocalScoreSearchAlgorithm
extends SearchAlgorithm

The ScoreBasedSearchAlgorithm class supports Bayes net structure search algorithms that are based on maximizing scores (as opposed to for example conditional independence based search algorithms).

Valid options are:

 -mbc
  Applies a Markov Blanket correction to the network structure, 
  after a network structure is learned. This ensures that all 
  nodes in the network are part of the Markov blanket of the 
  classifier node.

 -S [BAYES|MDL|ENTROPY|AIC|CROSS_CLASSIC|CROSS_BAYES]
  Score type (BAYES, BDeu, MDL, ENTROPY and AIC)

Version:

$Revision: 5196 $

Author:

Remco Bouckaert

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field and Description
static Tag[]	TAGS_SCORE_TYPE the score types

Constructor Summary

Constructors

Constructor and Description
LocalScoreSearchAlgorithm() default constructor
LocalScoreSearchAlgorithm(BayesNet bayesNet, Instances instances) constructor

Method Summary

Modifier and Type	Method and Description
void	buildStructure(BayesNet bayesNet, Instances instances) buildStructure determines the network structure/graph of the network with the K2 algorithm, restricted by its initial structure (which can be an empty graph, or a Naive Bayes graph.
double	calcNodeScore(int nNode) Calc Node Score for given parent set
double	calcScoreWithExtraParent(int nNode, int nCandidateParent) Calc Node Score With AddedParent
double	calcScoreWithMissingParent(int nNode, int nCandidateParent) Calc Node Score With Parent Deleted
boolean	getMarkovBlanketClassifier()
java.lang.String[]	getOptions() Gets the current settings of the search algorithm.
java.lang.String	getRevision() Returns the revision string.
SelectedTag	getScoreType() get quality measure to be used in searching for networks.
java.lang.String	globalInfo() This will return a string describing the search algorithm.
java.util.Enumeration	listOptions() Returns an enumeration describing the available options
double	logScore(int nType) logScore returns the log of the quality of a network (e.g.
java.lang.String	markovBlanketClassifierTipText()
java.lang.String	scoreTypeTipText()
void	setMarkovBlanketClassifier(boolean bMarkovBlanketClassifier)
void	setOptions(java.lang.String[] options) Parses a given list of options.
void	setScoreType(SelectedTag newScoreType) set quality measure to be used in searching for networks.

Methods inherited from class weka.classifiers.bayes.net.search.SearchAlgorithm

initAsNaiveBayesTipText, maxNrOfParentsTipText, toString

Methods inherited from class java.lang.Object

equals, getClass, hashCode, notify, notifyAll, wait, wait, wait

Field Detail

TAGS_SCORE_TYPE

public static final Tag[] TAGS_SCORE_TYPE

the score types

Constructor Detail

LocalScoreSearchAlgorithm

public LocalScoreSearchAlgorithm()

default constructor

LocalScoreSearchAlgorithm

public LocalScoreSearchAlgorithm(BayesNet bayesNet,
                                 Instances instances)

constructor

Parameters:

bayesNet - the network

instances - the data

Method Detail

logScore

public double logScore(int nType)

logScore returns the log of the quality of a network (e.g. the posterior probability of the network, or the MDL value).

Parameters:

nType - score type (Bayes, MDL, etc) to calculate score with

Returns:

log score.

buildStructure

public void buildStructure(BayesNet bayesNet,
                           Instances instances)
                    throws java.lang.Exception

buildStructure determines the network structure/graph of the network with the K2 algorithm, restricted by its initial structure (which can be an empty graph, or a Naive Bayes graph.

Overrides:

buildStructure in class SearchAlgorithm

Parameters:

bayesNet - the network

instances - the data to use

Throws:

java.lang.Exception - if something goes wrong

calcNodeScore

public double calcNodeScore(int nNode)

Calc Node Score for given parent set

Parameters:

nNode - node for which the score is calculate

Returns:

log score

calcScoreWithExtraParent

public double calcScoreWithExtraParent(int nNode,
                                       int nCandidateParent)

Calc Node Score With AddedParent

Parameters:

nNode - node for which the score is calculate

nCandidateParent - candidate parent to add to the existing parent set

Returns:

log score

calcScoreWithMissingParent

public double calcScoreWithMissingParent(int nNode,
                                         int nCandidateParent)

Calc Node Score With Parent Deleted

Parameters:

nNode - node for which the score is calculate

nCandidateParent - candidate parent to delete from the existing parent set

Returns:

log score

setScoreType

public void setScoreType(SelectedTag newScoreType)

set quality measure to be used in searching for networks.

Parameters:

newScoreType - the new score type

getScoreType

public SelectedTag getScoreType()

get quality measure to be used in searching for networks.

Returns:

quality measure

setMarkovBlanketClassifier

public void setMarkovBlanketClassifier(boolean bMarkovBlanketClassifier)

Parameters:

bMarkovBlanketClassifier -

getMarkovBlanketClassifier

public boolean getMarkovBlanketClassifier()

Returns:

listOptions

public java.util.Enumeration listOptions()

Returns an enumeration describing the available options

Specified by:

listOptions in interface OptionHandler

Overrides:

listOptions in class SearchAlgorithm

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:

 -mbc
  Applies a Markov Blanket correction to the network structure, 
  after a network structure is learned. This ensures that all 
  nodes in the network are part of the Markov blanket of the 
  classifier node.

 -S [BAYES|MDL|ENTROPY|AIC|CROSS_CLASSIC|CROSS_BAYES]
  Score type (BAYES, BDeu, MDL, ENTROPY and AIC)

Specified by:

setOptions in interface OptionHandler

Overrides:

setOptions in class SearchAlgorithm

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 search algorithm.

Specified by:

getOptions in interface OptionHandler

Overrides:

getOptions in class SearchAlgorithm

Returns:

an array of strings suitable for passing to setOptions

scoreTypeTipText

public java.lang.String scoreTypeTipText()

Returns:

a string to describe the ScoreType option.

markovBlanketClassifierTipText

public java.lang.String markovBlanketClassifierTipText()

Returns:

a string to describe the MarkovBlanketClassifier option.

globalInfo

public java.lang.String globalInfo()

This will return a string describing the search algorithm.

Returns:

The string.

getRevision

public java.lang.String getRevision()

Returns the revision string.

Specified by:

getRevision in interface RevisionHandler

Overrides:

getRevision in class SearchAlgorithm

Returns:

the revision