org.dllearner.learningproblems

Class FuzzyPosNegLPStandard

java.lang.Object

org.dllearner.core.AbstractComponent

org.dllearner.core.AbstractLearningProblem<T,org.semanticweb.owlapi.model.OWLClassExpression,EvaluatedDescription<T>>

org.dllearner.core.AbstractClassExpressionLearningProblem<ScorePosNeg<org.semanticweb.owlapi.model.OWLNamedIndividual>>

org.dllearner.learningproblems.FuzzyPosNegLP

org.dllearner.learningproblems.FuzzyPosNegLPStandard

All Implemented Interfaces:

Component, LearningProblem

public class FuzzyPosNegLPStandard
extends FuzzyPosNegLP

The aim of this learning problem is to learn a concept definition such that the positive examples and the negative examples do not follow. It is 2-valued, because we only distinguish between covered and non-covered examples. (A 3-valued problem distinguishes between covered examples, examples covered by the negation of the concept, and all other examples.) The 2-valued learning problem is often more useful for Description Logics due to (the Open World Assumption and) the fact that negative knowledge, e.g. that a person does not have a child, is or cannot be expressed.

Author:

Jens Lehmann

Constructor Summary

Constructors

Constructor and Description
FuzzyPosNegLPStandard()
FuzzyPosNegLPStandard(AbstractReasonerComponent reasoningService)
FuzzyPosNegLPStandard(AbstractReasonerComponent reasoningService, SortedSet<org.semanticweb.owlapi.model.OWLIndividual> positiveExamples, SortedSet<org.semanticweb.owlapi.model.OWLIndividual> negativeExamples)

Method Summary

Modifier and Type	Method and Description
ScorePosNeg<org.semanticweb.owlapi.model.OWLNamedIndividual>	computeScore(org.semanticweb.owlapi.model.OWLClassExpression concept, double noise) Computes the Score of a given hypothesis with respect to this learning problem.
EvaluatedDescription	evaluate(org.semanticweb.owlapi.model.OWLClassExpression description) Evaluates the hypothesis by computing the score and returning an evaluated hypothesis of the correct type (ClassLearningProblem returns EvaluatedDescriptionClass instead of generic EvaluatedDescription).
double	getAccuracy(int posAsPos, int posAsNeg, int negAsPos, int negAsNeg, double noise)
double	getAccuracyOrTooWeak(org.semanticweb.owlapi.model.OWLClassExpression description, double noise) This method computes the accuracy and returns -1 instead of the accuracy if the accuracy of the hypothesis is below the given threshold and the accuracy of all more special w.r.t.
double	getApproxDelta()
double	getFMeasureOrTooWeakApprox(org.semanticweb.owlapi.model.OWLClassExpression description, double noise) Deprecated.
Heuristics.HeuristicType	getHeuristic()
void	init() Method to be called after the component has been configured.
boolean	isUseApproximations()
void	setAccuracyMethod(Heuristics.HeuristicType accuracyMethod)
void	setApproxDelta(double approxDelta)
void	setHeuristic(Heuristics.HeuristicType heuristic)
void	setUseApproximations(boolean useApproximations)

Methods inherited from class org.dllearner.learningproblems.FuzzyPosNegLP

getFuzzyExamples, getNegativeExamples, getPercentPerLengthUnit, getPositiveExamples, setFuzzyExamples, setFuzzyExamples, setNegativeExamples, setPositiveExamples

Methods inherited from class org.dllearner.core.AbstractClassExpressionLearningProblem

changeReasonerComponent, getExampleLoaderHelper, getReasoningUtil, setExampleLoaderHelper, setReasoner, setReasoningUtil

Methods inherited from class org.dllearner.core.AbstractLearningProblem

computeScore, evaluate, getAccuracyOrTooWeak, getReasoner

Methods inherited from class org.dllearner.core.AbstractComponent

isInitialized

Methods inherited from class java.lang.Object

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

Constructor Detail

FuzzyPosNegLPStandard

public FuzzyPosNegLPStandard()

FuzzyPosNegLPStandard

public FuzzyPosNegLPStandard(AbstractReasonerComponent reasoningService)

FuzzyPosNegLPStandard

public FuzzyPosNegLPStandard(AbstractReasonerComponent reasoningService,
                             SortedSet<org.semanticweb.owlapi.model.OWLIndividual> positiveExamples,
                             SortedSet<org.semanticweb.owlapi.model.OWLIndividual> negativeExamples)

Method Detail

init

public void init()
          throws ComponentInitException

Description copied from interface: Component

Method to be called after the component has been configured. Implementation of components can overwrite this method to perform setup and initialisation tasks for this component.

Specified by:

init in interface Component

Overrides:

init in class FuzzyPosNegLP

Throws:

ComponentInitException - This exception is thrown if any exceptions occur within the initialisation process of this component. As component developer, you are encouraged to re-throw occurring exception as ComponentInitException and giving an error message as well as the actually exception by using the constructor ComponentInitException(String, Throwable).

getAccuracyOrTooWeak

public double getAccuracyOrTooWeak(org.semanticweb.owlapi.model.OWLClassExpression description,
                                   double noise)

Description copied from class: AbstractLearningProblem

This method computes the accuracy and returns -1 instead of the accuracy if

1. the accuracy of the hypothesis is below the given threshold and
2. the accuracy of all more special w.r.t. subsumption hypotheses is below the given threshold.

This is used for efficiency reasons, i.e. -1 can be returned instantly if it is clear that the hypothesis and all its refinements are not sufficiently accurate.

Specified by:

getAccuracyOrTooWeak in class AbstractLearningProblem<ScorePosNeg<org.semanticweb.owlapi.model.OWLNamedIndividual>,org.semanticweb.owlapi.model.OWLClassExpression,EvaluatedDescription<ScorePosNeg<org.semanticweb.owlapi.model.OWLNamedIndividual>>>

Returns:

A value between 0 and 1 indicating the quality (of a hypothesis) or -1 as described above.

getFMeasureOrTooWeakApprox

@Deprecated
public double getFMeasureOrTooWeakApprox(org.semanticweb.owlapi.model.OWLClassExpression description,
                                                     double noise)

Deprecated.

evaluate

public EvaluatedDescription evaluate(org.semanticweb.owlapi.model.OWLClassExpression description)

Description copied from class: AbstractLearningProblem

Evaluates the hypothesis by computing the score and returning an evaluated hypothesis of the correct type (ClassLearningProblem returns EvaluatedDescriptionClass instead of generic EvaluatedDescription).

Overrides:

evaluate in class AbstractLearningProblem<ScorePosNeg<org.semanticweb.owlapi.model.OWLNamedIndividual>,org.semanticweb.owlapi.model.OWLClassExpression,EvaluatedDescription<ScorePosNeg<org.semanticweb.owlapi.model.OWLNamedIndividual>>>

Parameters:

description - Hypothesis to evaluate.

Returns:

an evaluated hypothesis

getApproxDelta

public double getApproxDelta()

setApproxDelta

public void setApproxDelta(double approxDelta)

isUseApproximations

public boolean isUseApproximations()

setUseApproximations

public void setUseApproximations(boolean useApproximations)

getHeuristic

public Heuristics.HeuristicType getHeuristic()

setHeuristic

public void setHeuristic(Heuristics.HeuristicType heuristic)

setAccuracyMethod

public void setAccuracyMethod(Heuristics.HeuristicType accuracyMethod)

Parameters:

accuracyMethod - the accuracy method to set

getAccuracy

public double getAccuracy(int posAsPos,
                          int posAsNeg,
                          int negAsPos,
                          int negAsNeg,
                          double noise)

computeScore

public ScorePosNeg<org.semanticweb.owlapi.model.OWLNamedIndividual> computeScore(org.semanticweb.owlapi.model.OWLClassExpression concept,
                                                                                 double noise)

Description copied from class: AbstractLearningProblem

Computes the Score of a given hypothesis with respect to this learning problem. This can (but does not need to) be used by learning algorithms to measure how good the hypothesis fits the learning problem. Score objects are used to store e.g. covered examples, accuracy etc., so often it is more efficient to only create score objects for promising hypotheses.

Specified by:

computeScore in class AbstractLearningProblem<ScorePosNeg<org.semanticweb.owlapi.model.OWLNamedIndividual>,org.semanticweb.owlapi.model.OWLClassExpression,EvaluatedDescription<ScorePosNeg<org.semanticweb.owlapi.model.OWLNamedIndividual>>>

Parameters:

concept - A hypothesis (as solution candidate for this learning problem).

noise - the (approximated) value of noise within the examples

Returns:

A Score object.