Package org.dkpro.statistics.agreement.distance

Interface IDistanceFunction

All Known Implementing Classes:
CachedDistanceFunction, IntervalDistanceFunction, MASISetAnnotationDistanceFunction, NominalDistanceFunction, OrdinalDistanceFunction, RatioDistanceFunction, SetAnnotationDistanceFunction, SquareDistanceFunction
public interface IDistanceFunction
Interface for a distance function to be used in any IWeightedAgreement measure, such as WeightedKappaAgreement or KrippendorffAlphaAgreement. These measures use the distance function for quantifying the degree of disagreement between a pair of categories (i.e., two annotations assigned to a certain item by different raters). If annotation units are, for instance, coded using integers, the distance function might assign a smaller distance to the category pair (1, 2) then to the category pair (2, 9). See implementations of this interface for more detailed examples.
Author:
Christian M. Meyer
See Also:
IWeightedAgreement, WeightedKappaAgreement, KrippendorffAlphaAgreement

  • Method Summary

    Modifier and Type Method Description
    double measureDistance​(IAnnotationStudy study, java.lang.Object category1, java.lang.Object category2)
    Returns a distance value for the given pair of categories used within the given annotation study.

  • Method Details

    • measureDistance

      double measureDistance​(IAnnotationStudy study, java.lang.Object category1, java.lang.Object category2)
      Returns a distance value for the given pair of categories used within the given annotation study. Normally, the measure should return a distance of 0 if, and only if, the two categories are equal and a positive number otherwise.