Package de.jstacs.classifiers.performanceMeasures

This package provides the implementations of performance measures that can be used to assess any classifier.

See: Description

Interface Summary

Interface	Description
NumericalPerformanceMeasure	This interface indicates that a Performance measure returns numerical results.
PerformanceMeasure	Interface of any performance measure used to evaluate an AbstractClassifier.

Class Summary

Class	Description
AbstractNumericalTwoClassPerformanceMeasure	This class is the abstract super class of any performance measure that can only be computed for binary classifiers.
AbstractPerformanceMeasure	This class is the abstract super class of any performance measure used to evaluate an AbstractClassifier.
AbstractPerformanceMeasureParameterSet<T extends PerformanceMeasure>	This class implements a container of PerformanceMeasures that can be used in AbstractClassifier.evaluate(AbstractPerformanceMeasureParameterSet, boolean, de.jstacs.data.DataSet...).
AbstractTwoClassPerformanceMeasure	This class is the abstract super class of any performance measure that can only be computed for binary classifiers.
AucPR	This class implements the area under curve of the precision-recall curve.
AucROC	This class implements the area under curve of the Receiver Operating Characteristics curve.
ClassificationRate	This class implements the classification rate, i.e.
ConfusionMatrix	This class implements the performance measure confusion matrix.
CorrelationCoefficient	PerformanceMeasure using Pearson or Spearman correlation between prediction scores and weighted class labels.
FalsePositiveRateForFixedSensitivity	This class implements the false positive rate for a fixed sensitivity.
MaximumCorrelationCoefficient	This class implements the maximum of the correlation coefficient .
MaximumFMeasure	Computes the maximum of the general F-measure given a positive real parameter .
MaximumNumericalTwoClassMeasure	This class prepares everything for an easy implementation of a maximum of any numerical performance measure.
NumericalPerformanceMeasureParameterSet	This class implements a container for NumericalPerformanceMeasures that can be used, for instance, in an repeated assessment, (cf.
PerformanceMeasureParameterSet	This class implements a container of AbstractPerformanceMeasures that can be used in AbstractClassifier.evaluate(AbstractPerformanceMeasureParameterSet, boolean, de.jstacs.data.DataSet...).
PositivePredictiveValueForFixedSensitivity	This class implements the positive predictive value for a fixed sensitivity.
PRCurve	This class implements the precision-recall curve and its area under the curve.
ROCCurve	This class implements the Receiver Operating Characteristics curve and the area under the curve.
SensitivityForFixedSpecificity	This class implements the sensitivity for a fixed specificity.

Enum Summary

Enum	Description
CorrelationCoefficient.Method	The type of correlation used.

Package de.jstacs.classifiers.performanceMeasures Description

This package provides the implementations of performance measures that can be used to assess any classifier.

See Also:

AbstractPerformanceMeasure, PerformanceMeasureParameterSet, AbstractClassifier