Package elki.distance

Interface Distance<O>

Type Parameters:

O - Object type

All Known Subinterfaces:

ClusteringDistanceSimilarity, DBIDDistance, DBIDRangeDistance, DimensionSelectingSubspaceDistance<O>, IndexBasedDistance<O>, Norm<O>, NumberVectorDistance<O>, PrimitiveDistance<O>, SpatialPrimitiveDistance<V>, WeightedNumberVectorDistance<V>

All Known Implementing Classes:

AbsolutePearsonCorrelationDistance, AbsoluteUncenteredCorrelationDistance, AbstractDatabaseDistance, AbstractDBIDRangeDistance, AbstractDimensionsSelectingDistance, AbstractEditDistance, AbstractIndexBasedDistance, AbstractNumberVectorDistance, AbstractSetDistance, AbstractSimilarityAdapter, ArcCosineDistance, ArcCosineUnitlengthDistance, ArccosSimilarityAdapter, BrayCurtisDistance, CanberraDistance, ChiDistance, ChiSquaredDistance, ClarkDistance, ClusteringAdjustedRandIndexSimilarity, ClusteringBCubedF1Similarity, ClusteringFowlkesMallowsSimilarity, ClusteringRandIndexSimilarity, ClusterIntersectionSimilarity, ClusterJaccardSimilarity, CosineDistance, CosineUnitlengthDistance, DerivativeDTWDistance, DimensionSelectingLatLngDistance, DiskCacheBasedDoubleDistance, DiskCacheBasedFloatDistance, DTWDistance, EDRDistance, ERPDistance, EuclideanDistance, FileBasedSparseDoubleDistance, FileBasedSparseFloatDistance, FisherRaoDistance, HammingDistance, HellingerDistance, HistogramIntersectionDistance, HistogramMatchDistance, HSBHistogramQuadraticDistance, JaccardSimilarityDistance, JeffreyDivergenceDistance, JensenShannonDivergenceDistance, KolmogorovSmirnovDistance, Kulczynski1Similarity, KullbackLeiblerDivergenceAsymmetricDistance, KullbackLeiblerDivergenceReverseAsymmetricDistance, LatLngDistance, LCSSDistance, LevenshteinDistance, LinearKernel, LinearSimilarityAdapter, LngLatDistance, LnSimilarityAdapter, LPIntegerNormDistance, LPNormDistance, MahalanobisDistance, ManhattanDistance, MatrixWeightedQuadraticDistance, MaximumDistance, MinimumDistance, MultiLPNorm, NormalizedLevenshteinDistance, OnedimensionalDistance, PearsonCorrelationDistance, PolynomialKernel, RandomStableDistance, RGBHistogramQuadraticDistance, SharedNearestNeighborJaccardDistance, SparseEuclideanDistance, SparseLPNormDistance, SparseManhattanDistance, SparseMaximumDistance, SparseSquaredEuclideanDistance, SqrtCosineDistance, SqrtCosineUnitlengthDistance, SqrtJensenShannonDivergenceDistance, SquaredEuclideanDistance, SquaredPearsonCorrelationDistance, SquaredUncenteredCorrelationDistance, SubspaceEuclideanDistance, SubspaceLPNormDistance, SubspaceManhattanDistance, SubspaceMaximumDistance, TriangularDiscriminationDistance, TriangularDistance, TutorialDistance, UncenteredCorrelationDistance, WeightedCanberraDistance, WeightedEuclideanDistance, WeightedLPNormDistance, WeightedManhattanDistance, WeightedMaximumDistance, WeightedPearsonCorrelationDistance, WeightedSquaredEuclideanDistance, WeightedSquaredPearsonCorrelationDistance

public interface Distance<O>

Base interface for any kind of distances.

Since:

0.1

Author:

Erich Schubert

Method Summary

Modifier and Type	Method	Description
TypeInformation	getInputTypeRestriction()	Get the input data type of the function.
<T extends O> DistanceQuery<T>	instantiate(Relation<T> relation)	Instantiate with a database to get the actual distance query.
default boolean	isMetric()	Is this distance function metric (satisfy the triangle inequality)
default boolean	isSquared()	Squared distances, that would become metric after square root.
default boolean	isSymmetric()	Is this function symmetric?

Method Detail

isSymmetric

default boolean isSymmetric()

Is this function symmetric?

Returns:

true when symmetric

isMetric

default boolean isMetric()

Is this distance function metric (satisfy the triangle inequality)

Returns:

true when metric.

isSquared

default boolean isSquared()

Squared distances, that would become metric after square root.

E.g. squared Euclidean.

Returns:

true when squared.

getInputTypeRestriction

TypeInformation getInputTypeRestriction()

Get the input data type of the function.

Returns:

Type restriction

instantiate

<T extends O> DistanceQuery<T> instantiate(Relation<T> relation)

Instantiate with a database to get the actual distance query.

Parameters:

relation - The representation to use

Returns:

Actual distance query.