org.aksw.limes.core.evaluation.qualititativeMeasures

Class PseudoFMeasure

java.lang.Object

org.aksw.limes.core.evaluation.qualititativeMeasures.APRF

org.aksw.limes.core.evaluation.qualititativeMeasures.APseudoPRF

org.aksw.limes.core.evaluation.qualititativeMeasures.PseudoFMeasure

All Implemented Interfaces:

IQualitativeMeasure

Direct Known Subclasses:

PseudoRefFMeasure

public class PseudoFMeasure
extends APseudoPRF

Implements a quality measure for unsupervised ML algorihtms, dubbed pseudo F-Measure.
Thereby, not relying on any gold standard. The basic idea is to measure the quality of the given Mapping by calculating how close it is to an assumed 1-to-1 Mapping between source and target.

Version:

1.0

Author:

Klaus Lyko (lyko@informatik.uni-leipzig.de), Axel-C. Ngonga Ngomo (ngonga@informatik.uni-leipzig.de), Mofeed Hassan (mounir@informatik.uni-leipzig.de)

Field Summary

Fields inherited from class org.aksw.limes.core.evaluation.qualititativeMeasures.APseudoPRF

symmetricPrecision

Constructor Summary

Constructors

Constructor and Description
PseudoFMeasure()
PseudoFMeasure(boolean symmetricPrecision) Use this constructor to toggle between symmetric precision (true) and the older asymmetric Pseudo-Precision (false)

Method Summary

Modifier and Type	Method and Description
double	calculate(AMapping predictions, GoldStandard goldStandard) The method calculates the pseudo F-Measure of the machine learning predictions compared to a gold standard for beta = 1 .
double	getPseudoFMeasure(AMapping predictions, GoldStandard goldStandard, double beta) The method calculates the pseudo F-Measure of the machine learning predictions compared to a gold standard for different beta values
double	precision(AMapping predictions, GoldStandard goldStandard) The method calculates the pseudo precision of the machine learning predictions compared to a gold standard
double	recall(AMapping predictions, GoldStandard goldStandard) The method calculates the pseudo recall of the machine learning predictions compared to a gold standard

Methods inherited from class org.aksw.limes.core.evaluation.qualititativeMeasures.APseudoPRF

getUse1To1Mapping, isSymmetricPrecision, isUse1To1Mapping, setSymmetricPrecision, setUse1To1Mapping

Methods inherited from class org.aksw.limes.core.evaluation.qualititativeMeasures.APRF

falseNegative, trueFalsePositive, trueNegative

Methods inherited from class java.lang.Object

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

Constructor Detail

PseudoFMeasure

public PseudoFMeasure()

PseudoFMeasure

public PseudoFMeasure(boolean symmetricPrecision)

Use this constructor to toggle between symmetric precision (true) and the older asymmetric Pseudo-Precision (false)

Parameters:

symmetricPrecision - sets/resets the symmetric precision flag

Method Detail

calculate

public double calculate(AMapping predictions,
                        GoldStandard goldStandard)

The method calculates the pseudo F-Measure of the machine learning predictions compared to a gold standard for beta = 1 .

Specified by:

calculate in interface IQualitativeMeasure

Specified by:

calculate in class APseudoPRF

Parameters:

predictions - The predictions provided by a machine learning algorithm.

goldStandard - It contains the gold standard (reference mapping) combined with the source and target URIs.

Returns:

double - This returns the calculated pseudo F-Measure.

getPseudoFMeasure

public double getPseudoFMeasure(AMapping predictions,
                                GoldStandard goldStandard,
                                double beta)

The method calculates the pseudo F-Measure of the machine learning predictions compared to a gold standard for different beta values

Parameters:

predictions - The predictions provided by a machine learning algorithm

goldStandard - It contains the gold standard (reference mapping) combined with the source and target URIs

beta - Beta for F-beta

Returns:

double - This returns the calculated pseudo F-Measure

recall

public double recall(AMapping predictions,
                     GoldStandard goldStandard)

The method calculates the pseudo recall of the machine learning predictions compared to a gold standard

Parameters:

predictions - The predictions provided by a machine learning algorithm

goldStandard - It contains the gold standard (reference mapping) combined with the source and target URIs

Returns:

double - This returns the calculated pseudo recall

precision

public double precision(AMapping predictions,
                        GoldStandard goldStandard)

The method calculates the pseudo precision of the machine learning predictions compared to a gold standard

Parameters:

predictions - The predictions provided by a machine learning algorithm

goldStandard - It contains the gold standard (reference mapping) combined with the source and target URIs

Returns:

double - This returns the calculated pseudo precision