org.dllearner.learningproblems

Class ScoreThreeValued<T extends org.semanticweb.owlapi.model.OWLEntity>

java.lang.Object

org.dllearner.core.Score

org.dllearner.learningproblems.ScorePosNeg<T>

org.dllearner.learningproblems.ScoreThreeValued<T>

All Implemented Interfaces:

Serializable

public class ScoreThreeValued<T extends org.semanticweb.owlapi.model.OWLEntity>
extends ScorePosNeg<T>

Computes the score (a negative value) by comparing the classification results with ideal results. TODO: The implementation is not very efficient, because some things are only computed to be able to present the score results. This means that it would be better to compute only the necessary computations and do the other ones only when they are needed to calculate statistical values.

Author:

Jens Lehmann

See Also:

Serialized Form

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	ScoreThreeValued.ScoreMethod

Constructor Summary

Constructors

Constructor and Description
ScoreThreeValued(int conceptLength, double accuracyPenalty, double errorPenalty, boolean penaliseNeutralExamples, double percentPerLengthUnit, Set<T> posClassified, Set<T> neutClassified, Set<T> negClassified, Set<T> posExamples, Set<T> neutExamples, Set<T> negExamples)

Method Summary

Modifier and Type	Method and Description
double	getAccuracy() This method returns a value, which indicates how accurate a hypothesis solves a learning problem.
int	getCommissionErrorRate()
Set<T>	getCoveredNegatives()
Set<T>	getCoveredPositives()
int	getInductionRate()
int	getMatchRate()
ScorePosNeg<T>	getModifiedLengthScore(int newLength) The score of a concept depends on how good it classifies the examples of a learning problem and the length of the concept itself.
Set<T>	getNegClassified()
Set<T>	getNotCoveredNegatives()
Set<T>	getNotCoveredPositives()
int	getOmmissionErrorRate()
Set<T>	getPosClassified()
double	getScoreValue()
String	toString()

Methods inherited from class java.lang.Object

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

Constructor Detail

ScoreThreeValued

public ScoreThreeValued(int conceptLength,
                        double accuracyPenalty,
                        double errorPenalty,
                        boolean penaliseNeutralExamples,
                        double percentPerLengthUnit,
                        Set<T> posClassified,
                        Set<T> neutClassified,
                        Set<T> negClassified,
                        Set<T> posExamples,
                        Set<T> neutExamples,
                        Set<T> negExamples)

Method Detail

getScoreValue

public double getScoreValue()

Specified by:

getScoreValue in class ScorePosNeg<T extends org.semanticweb.owlapi.model.OWLEntity>

getMatchRate

public int getMatchRate()

Returns:

number of cases of individuals that got exactly the same classification with both definitions

getOmmissionErrorRate

public int getOmmissionErrorRate()

Returns:

amount of individuals for which class-membership w.r.t. the given query could not determined using the induced definition, while they actually belong (do not belong) to the query concept

getCommissionErrorRate

public int getCommissionErrorRate()

Returns:

amount of individuals found not to belong to the query concept according to the induced definition, while they actually belong to it and vice-versa

getInductionRate

public int getInductionRate()

Returns:

amount of individuals found to belong or not to belong to the query concept according to the induced definition, while either case is not logically derivable from the knowledge base with the original definition

toString

public String toString()

Overrides:

toString in class Object

getNegClassified

public Set<T> getNegClassified()

getPosClassified

public Set<T> getPosClassified()

getCoveredNegatives

public Set<T> getCoveredNegatives()

Specified by:

getCoveredNegatives in class ScorePosNeg<T extends org.semanticweb.owlapi.model.OWLEntity>

getCoveredPositives

public Set<T> getCoveredPositives()

Specified by:

getCoveredPositives in class ScorePosNeg<T extends org.semanticweb.owlapi.model.OWLEntity>

getNotCoveredPositives

public Set<T> getNotCoveredPositives()

Specified by:

getNotCoveredPositives in class ScorePosNeg<T extends org.semanticweb.owlapi.model.OWLEntity>

getNotCoveredNegatives

public Set<T> getNotCoveredNegatives()

Specified by:

getNotCoveredNegatives in class ScorePosNeg<T extends org.semanticweb.owlapi.model.OWLEntity>

getModifiedLengthScore

public ScorePosNeg<T> getModifiedLengthScore(int newLength)

Description copied from class: ScorePosNeg

The score of a concept depends on how good it classifies the examples of a learning problem and the length of the concept itself. If a given concept is known to have equal classification properties than the concept this score object is based on, then this method can be used to calculate its score value by using the length of this concept as parameter.

Specified by:

getModifiedLengthScore in class ScorePosNeg<T extends org.semanticweb.owlapi.model.OWLEntity>

Parameters:

newLength - Length of the concept.

Returns:

Score.

getAccuracy

public double getAccuracy()

Description copied from class: Score

This method returns a value, which indicates how accurate a hypothesis solves a learning problem.

Specified by:

getAccuracy in class Score

Returns:

A value between 0 and 1 indicating the quality (of a hypothesis).

See Also:

AbstractLearningProblem.getAccuracyOrTooWeak(OWLObject, double)