mulan.evaluation.measure
Class LabelBasedFMeasure

java.lang.Object
  mulan.evaluation.measure.MeasureBase
      mulan.evaluation.measure.BipartitionMeasureBase
          mulan.evaluation.measure.LabelBasedBipartitionMeasureBase
              mulan.evaluation.measure.LabelBasedFMeasure

All Implemented Interfaces:

Serializable, Measure

Direct Known Subclasses:

MacroFMeasure, MicroFMeasure

public abstract class LabelBasedFMeasure
extends LabelBasedBipartitionMeasureBase

Base implementation of the label-based macro/micro f-measures.

Version:

2010.12.31

Author:

Grigorios Tsoumakas

See Also:

Serialized Form

Field Summary

protected double

beta the parameter for combining precision and recall

Fields inherited from class mulan.evaluation.measure.LabelBasedBipartitionMeasureBase

falseNegatives, falsePositives, numOfLabels, trueNegatives, truePositives

Constructor Summary

LabelBasedFMeasure(int numOfLabels)
Constructs a new object with given number of labels

LabelBasedFMeasure(int numOfLabels, double beta)
Constructs a new object with given number of labels and beta parameter

Method Summary

double

getIdealValue() Gets an 'ideal' value of a measure.

Methods inherited from class mulan.evaluation.measure.LabelBasedBipartitionMeasureBase

reset, updateBipartition

Methods inherited from class mulan.evaluation.measure.BipartitionMeasureBase

updateInternal

Methods inherited from class mulan.evaluation.measure.MeasureBase

makeCopy, toString, update

Methods inherited from class java.lang.Object

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

Methods inherited from interface mulan.evaluation.measure.Measure

getName, getValue

Field Detail

beta

protected final double beta

the parameter for combining precision and recall

Constructor Detail

LabelBasedFMeasure

public LabelBasedFMeasure(int numOfLabels)

Constructs a new object with given number of labels

Parameters:

numOfLabels - the number of labels

LabelBasedFMeasure

public LabelBasedFMeasure(int numOfLabels,
                          double beta)

Constructs a new object with given number of labels and beta parameter

Parameters:

numOfLabels - the number of labels

beta - the beta parameter

Method Detail

getIdealValue

public double getIdealValue()

Description copied from interface: Measure

Gets an 'ideal' value of a measure. The 'ideal' means, that the value represents the best achievable performance of a learner for an prediction of a multi-label task and associated true labels.

Returns:

the ideal value