Similarity (S-Space Package 2.0.1 API)

java.lang.Object
- edu.ucla.sspace.common.Similarity

```
public class Similarity
extends Object
```
A collection of static methods for computing the similarity, distances, correlations, and all other vectorized value comparisons between vectors and arrays. This is the primary class for comparing the term and document vectors produced by SemanticSpace.

Author:

Keith Stevens, David Jurgens

Nested Class Summary

Nested Classes
Modifier and Type Class and Description

static class Similarity.SimType
A type of similarity function to use when generating a Method

Nested Classes
Modifier and Type	Class and Description
`static class`	`Similarity.SimType` A type of similarity function to use when generating a `Method`

Method Summary

Methods
Modifier and Type	Method and Description
`static double`	`averageCommonFeatureRank(double[] a, double[] b)` Computes the Average Common Feature Rank between the two feature arrays.
`static double`	`averageCommonFeatureRank(int[] a, int[] b)` Computes the Average Common Feature Rank between the two feature arrays.
`static double`	`averageCommonFeatureRank(Vector a, Vector b)` Computes the Average Common Feature Rank between the two feature arrays.
`static double`	`correlation(double[] arr1, double[] arr2)` Returns the Pearson product-moment correlation coefficient of the two arrays.
`static double`	`correlation(DoubleVector arr1, DoubleVector arr2)` Returns the Pearson product-moment correlation coefficient of the two `Vector`s.
`static double`	`correlation(int[] arr1, int[] arr2)` Returns the Pearson product-moment correlation coefficient of the two arrays.
`static double`	`correlation(IntegerVector arr1, DoubleVector arr2)` Returns the Pearson product-moment correlation coefficient of the two `Vector`s.
`static double`	`correlation(Vector a, Vector b)` Returns the Pearson product-moment correlation coefficient of the two `Vector`s.
`static double`	`cosineSimilarity(double[] a, double[] b)` Returns the cosine similarity of the two arrays.
`static double`	`cosineSimilarity(DoubleVector a, DoubleVector b)` Returns the cosine similarity of the two `DoubleVector`.
`static double`	`cosineSimilarity(int[] a, int[] b)` Returns the cosine similarity of the two arrays.
`static double`	`cosineSimilarity(IntegerVector a, IntegerVector b)` Returns the cosine similarity of the two `DoubleVector`.
`static double`	`cosineSimilarity(Vector a, Vector b)` Returns the cosine similarity of the two `DoubleVector`.
`static double`	`euclideanDistance(double[] a, double[] b)` Returns the euclidian distance between two arrays of {code double}s.
`static double`	`euclideanDistance(DoubleVector a, DoubleVector b)` Returns the euclidian distance between two `DoubleVector`s.
`static double`	`euclideanDistance(int[] a, int[] b)` Returns the euclidian distance between two arrays of {code double}s.
`static double`	`euclideanDistance(IntegerVector a, IntegerVector b)` Returns the euclidian distance between two `DoubleVector`s.
`static double`	`euclideanDistance(Vector a, Vector b)` Returns the euclidian distance between two `Vector`s.
`static double`	`euclideanSimilarity(double[] a, double[] b)` Returns the euclidian similiarty between two arrays of values.
`static double`	`euclideanSimilarity(int[] a, int[] b)` Returns the euclidian similiarty between two arrays of values.
`static double`	`euclideanSimilarity(Vector a, Vector b)` Returns the euclidian similiarty between two arrays of values.
`static Method`	`getMethod(Similarity.SimType similarityType)` Deprecated. The `getSimilarity(SimType,double[],double[])` method should be used instead. Returns the `Method` for getting the similarity of two `double[]` based on the specified similarity type.
`static double`	`getSimilarity(Similarity.SimType similarityType, double[] a, double[] b)` Calculates the similarity of the two vectors using the provided similarity measure.
`static <T extends Vector> double`	`getSimilarity(Similarity.SimType similarityType, T a, T b)` Calculates the similarity of the two vectors using the provided similarity measure.
`static SimilarityFunction`	`getSimilarityFunction(Similarity.SimType similarityType)`
`static double`	`goodmanKruskalGamma(DoubleVector a, DoubleVector b)` Computes the Goodman-Kruskal Gamma coefficient, which is preferrable to Kendall's τ and Spearman's ρ when there are many ties in the data.
`static double`	`jaccardIndex(double[] a, double[] b)` Computes the Jaccard index comparing the similarity both arrays when viewed as sets of samples.
`static double`	`jaccardIndex(DoubleVector a, DoubleVector b)` Computes the Jaccard index comparing the similarity both `DoubleVector`s when viewed as sets of samples.
`static double`	`jaccardIndex(int[] a, int[] b)` Computes the Jaccard index comparing the similarity both arrays when viewed as sets of samples.
`static double`	`jaccardIndex(IntegerVector a, IntegerVector b)` Computes the Jaccard index comparing the similarity both `IntegerVector`s when viewed as sets of samples.
`static double`	`jaccardIndex(Set<?> a, Set<?> b)` Computes the Jaccard index of the two sets of elements.
`static double`	`jaccardIndex(Vector a, Vector b)` Computes the Jaccard index comparing the similarity both `Vector`s when viewed as sets of samples.
`static double`	`kendallsTau(double[] a, double[] b)` Computes Kendall's tau of the values in the two arrays.
`static double`	`kendallsTau(DoubleVector a, DoubleVector b)` Computes Kendall's tau of the values in the two vectors.
`static double`	`kendallsTau(int[] a, int[] b)` Computes Kendall's tau of the values in the two arrays.
`static double`	`kendallsTau(IntegerVector a, IntegerVector b)` Computes Kendall's tau of the values in the two vectors.
`static double`	`kendallsTau(Vector a, Vector b)` Computes Kendall's tau of the values in the two vectors.
`static double`	`klDivergence(double[] a, double[] b)` Computes the K-L Divergence of two probability distributions `A` and `B` where the vectors `a` and `b` correspond to `n` samples from each respective distribution.
`static double`	`klDivergence(DoubleVector a, DoubleVector b)` Computes the K-L Divergence of two probability distributions `A` and `B` where the vectors `a` and `b` correspond to `n` samples from each respective distribution.
`static double`	`klDivergence(int[] a, int[] b)` Computes the K-L Divergence of two probability distributions `A` and `B` where the vectors `a` and `b` correspond to `n` samples from each respective distribution.
`static double`	`klDivergence(IntegerVector a, IntegerVector b)` Computes the K-L Divergence of two probability distributions `A` and `B` where the vectors `a` and `b` correspond to `n` samples from each respective distribution.
`static double`	`klDivergence(Vector a, Vector b)` Computes the K-L Divergence of two probability distributions `A` and `B` where the vectors `a` and `b` correspond to `n` samples from each respective distribution.
`static double`	`linSimilarity(double[] a, double[] b)` Computes the lin similarity measure, which is motivated by information theory priniciples.
`static double`	`linSimilarity(DoubleVector a, DoubleVector b)` Computes the lin similarity measure, which is motivated by information theory priniciples.
`static double`	`linSimilarity(int[] a, int[] b)` Computes the lin similarity measure, which is motivated by information theory priniciples.
`static double`	`linSimilarity(IntegerVector a, IntegerVector b)` Computes the lin similarity measure, which is motivated by information theory priniciples.
`static double`	`linSimilarity(Vector a, Vector b)` Computes the lin similarity measure, which is motivated by information theory priniciples.
`static double`	`spearmanRankCorrelationCoefficient(double[] a, double[] b)` Computes the Spearman rank correlation coefficient for the two arrays.
`static double`	`spearmanRankCorrelationCoefficient(DoubleVector a, DoubleVector b)` Computes the Spearman rank correlation coefficient for the two `DoubleVector`s.
`static double`	`spearmanRankCorrelationCoefficient(int[] a, int[] b)` Computes the Spearman rank correlation coefficient for the two arrays.
`static double`	`spearmanRankCorrelationCoefficient(IntegerVector a, IntegerVector b)` Computes the Spearman rank correlation coefficient for the two `DoubleVector`s.
`static double`	`spearmanRankCorrelationCoefficient(Vector a, Vector b)` Computes the Spearman rank correlation coefficient for the two `Vector`s.
`static double`	`tanimotoCoefficient(double[] a, double[] b)` Returns the Tanimoto coefficient of the two `double` array instances.
`static double`	`tanimotoCoefficient(DoubleVector a, DoubleVector b)` Returns the Tanimoto coefficient of the two `DoubleVector` instances.
`static double`	`tanimotoCoefficient(int[] a, int[] b)` Returns the Tanimoto coefficient of the two `int` array instances.
`static double`	`tanimotoCoefficient(IntegerVector a, IntegerVector b)` Returns the Tanimoto Coefficient of the two `IntegerVector` instances
`static double`	`tanimotoCoefficient(Vector a, Vector b)` Returns the Tanimoto coefficient of the two `Vector` instances.

Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

- Method Detail
  - getMethod
```
@Deprecated
public static Method getMethod(Similarity.SimType similarityType)
```
    Deprecated. The getSimilarity(SimType,double[],double[]) method should be used instead. Returns the Method for getting the similarity of two double[] based on the specified similarity type.
    
    Throws:
    
    Error - if a NoSuchMethodException is thrown
  - getSimilarityFunction
```
public static SimilarityFunction getSimilarityFunction(Similarity.SimType similarityType)
```
  - getSimilarity
```
public static double getSimilarity(Similarity.SimType similarityType,
                   double[] a,
                   double[] b)
```
    Calculates the similarity of the two vectors using the provided similarity measure.
    
    Parameters:
    similarityType - the similarity evaluation to use when comparing a and b
    a - a vector
    b - a vector
    
    Returns:
    the similarity according to the specified measure
  - getSimilarity
```
public static <T extends Vector> double getSimilarity(Similarity.SimType similarityType,
                                      T a,
                                      T b)
```
    Calculates the similarity of the two vectors using the provided similarity measure.
    
    Parameters:
    similarityType - the similarity evaluation to use when comparing a and b
    a - a Vector
    b - a Vector
    
    Returns:
    the similarity according to the specified measure
  - cosineSimilarity
```
public static double cosineSimilarity(double[] a,
                      double[] b)
```
    Returns the cosine similarity of the two arrays.
    
    Throws:
    
    IllegalArgumentException - when the length of the two vectors are not the same.
  - cosineSimilarity
```
public static double cosineSimilarity(int[] a,
                      int[] b)
```
    Returns the cosine similarity of the two arrays.
    
    Throws:
    
    IllegalArgumentException - when the length of the two vectors are not the same.
  - cosineSimilarity
```
public static double cosineSimilarity(DoubleVector a,
                      DoubleVector b)
```
    Returns the cosine similarity of the two DoubleVector.
  - cosineSimilarity
```
public static double cosineSimilarity(IntegerVector a,
                      IntegerVector b)
```
    Returns the cosine similarity of the two DoubleVector.
    
    Throws:
    
    IllegalArgumentException - when the length of the two vectors are not the same.
  - cosineSimilarity
```
public static double cosineSimilarity(Vector a,
                      Vector b)
```
    Returns the cosine similarity of the two DoubleVector.
    
    Throws:
    
    IllegalArgumentException - when the length of the two vectors are not the same.
  - correlation
```
public static double correlation(double[] arr1,
                 double[] arr2)
```
    Returns the Pearson product-moment correlation coefficient of the two arrays.
    
    Throws:
    
    IllegalArgumentException - when the length of the two vectors are not the same.
  - correlation
```
public static double correlation(int[] arr1,
                 int[] arr2)
```
    Returns the Pearson product-moment correlation coefficient of the two arrays.
    
    Throws:
    
    IllegalArgumentException - when the length of the two vectors are not the same.
  - correlation
```
public static double correlation(DoubleVector arr1,
                 DoubleVector arr2)
```
    Returns the Pearson product-moment correlation coefficient of the two Vectors.
    
    Throws:
    
    IllegalArgumentException - when the length of the two vectors are not the same.
  - correlation
```
public static double correlation(IntegerVector arr1,
                 DoubleVector arr2)
```
    Returns the Pearson product-moment correlation coefficient of the two Vectors.
    
    Throws:
    
    IllegalArgumentException - when the length of the two vectors are not the same.
  - correlation
```
public static double correlation(Vector a,
                 Vector b)
```
    Returns the Pearson product-moment correlation coefficient of the two Vectors.
    
    Throws:
    
    IllegalArgumentException - when the length of the two vectors are not the same.
  - euclideanDistance
```
public static double euclideanDistance(double[] a,
                       double[] b)
```
    Returns the euclidian distance between two arrays of {code double}s.
    
    Throws:
    
    IllegalArgumentException - when the length of the two vectors are not the same.
  - euclideanDistance
```
public static double euclideanDistance(int[] a,
                       int[] b)
```
    Returns the euclidian distance between two arrays of {code double}s.
    
    Throws:
    
    IllegalArgumentException - when the length of the two vectors are not the same.
  - euclideanDistance
```
public static double euclideanDistance(DoubleVector a,
                       DoubleVector b)
```
    Returns the euclidian distance between two DoubleVectors.
    
    Throws:
    
    IllegalArgumentException - when the length of the two vectors are not the same.
  - euclideanDistance
```
public static double euclideanDistance(IntegerVector a,
                       IntegerVector b)
```
    Returns the euclidian distance between two DoubleVectors.
    
    Throws:
    
    IllegalArgumentException - when the length of the two vectors are not the same.
  - euclideanDistance
```
public static double euclideanDistance(Vector a,
                       Vector b)
```
    Returns the euclidian distance between two Vectors.
    
    Throws:
    
    IllegalArgumentException - when the length of the two vectors are not the same.
  - euclideanSimilarity
```
public static double euclideanSimilarity(int[] a,
                         int[] b)
```
    Returns the euclidian similiarty between two arrays of values.
    
    Throws:
    
    IllegalArgumentException - when the length of the two vectors are not the same.
  - euclideanSimilarity
```
public static double euclideanSimilarity(double[] a,
                         double[] b)
```
    Returns the euclidian similiarty between two arrays of values.
    
    Throws:
    
    IllegalArgumentException - when the length of the two vectors are not the same.
  - euclideanSimilarity
```
public static double euclideanSimilarity(Vector a,
                         Vector b)
```
    Returns the euclidian similiarty between two arrays of values.
    
    Throws:
    
    IllegalArgumentException - when the length of the two vectors are not the same.
  - jaccardIndex
```
public static double jaccardIndex(Set<?> a,
                  Set<?> b)
```
    Computes the Jaccard index of the two sets of elements.
  - jaccardIndex
```
public static double jaccardIndex(double[] a,
                  double[] b)
```
    Computes the Jaccard index comparing the similarity both arrays when viewed as sets of samples.
  - jaccardIndex
```
public static double jaccardIndex(int[] a,
                  int[] b)
```
    Computes the Jaccard index comparing the similarity both arrays when viewed as sets of samples.
  - jaccardIndex
```
public static double jaccardIndex(DoubleVector a,
                  DoubleVector b)
```
    Computes the Jaccard index comparing the similarity both DoubleVectors when viewed as sets of samples.
  - jaccardIndex
```
public static double jaccardIndex(IntegerVector a,
                  IntegerVector b)
```
    Computes the Jaccard index comparing the similarity both IntegerVectors when viewed as sets of samples.
  - jaccardIndex
```
public static double jaccardIndex(Vector a,
                  Vector b)
```
    Computes the Jaccard index comparing the similarity both Vectors when viewed as sets of samples.
  - spearmanRankCorrelationCoefficient
```
public static double spearmanRankCorrelationCoefficient(double[] a,
                                        double[] b)
```
    Computes the Spearman rank correlation coefficient for the two arrays. If there is a tie in the ranking of a, then Pearson's product-moment coefficient is returned instead.
    
    Throws:
    
    IllegalArgumentException - when the length of the two vectors are not the same.
  - spearmanRankCorrelationCoefficient
```
public static double spearmanRankCorrelationCoefficient(int[] a,
                                        int[] b)
```
    Computes the Spearman rank correlation coefficient for the two arrays. If there is a tie in the ranking of a, then Pearson's product-moment coefficient is returned instead.
    
    Throws:
    
    IllegalArgumentException - when the length of the two vectors are not the same.
  - spearmanRankCorrelationCoefficient
```
public static double spearmanRankCorrelationCoefficient(DoubleVector a,
                                        DoubleVector b)
```
    Computes the Spearman rank correlation coefficient for the two DoubleVectors. If there is a tie in the ranking of a, then Pearson's product-moment coefficient is returned instead.
    
    Throws:
    
    IllegalArgumentException - when the length of the two vectors are not the same.
    
    IllegalArgumentException - when the length of the two vectors are not the same.
  - spearmanRankCorrelationCoefficient
```
public static double spearmanRankCorrelationCoefficient(IntegerVector a,
                                        IntegerVector b)
```
    Computes the Spearman rank correlation coefficient for the two DoubleVectors. If there is a tie in the ranking of a, then Pearson's product-moment coefficient is returned instead.
    
    Throws:
    
    IllegalArgumentException - when the length of the two vectors are not the same.
  - spearmanRankCorrelationCoefficient
```
public static double spearmanRankCorrelationCoefficient(Vector a,
                                        Vector b)
```
    Computes the Spearman rank correlation coefficient for the two Vectors. If there is a tie in the ranking of a, then Pearson's product-moment coefficient is returned instead.
    
    Throws:
    
    IllegalArgumentException - when the length of the two vectors are not the same.
  - averageCommonFeatureRank
```
public static double averageCommonFeatureRank(double[] a,
                              double[] b)
```
    Computes the Average Common Feature Rank between the two feature arrays. Uses the top 20 features for comparison.
    
    Throws:
    
    IllegalArgumentException - when the length of the two vectors are not the same.
  - averageCommonFeatureRank
```
public static double averageCommonFeatureRank(Vector a,
                              Vector b)
```
    Computes the Average Common Feature Rank between the two feature arrays. Uses the top 20 features for comparison. Converts types and calls averageCommonFeatureRank(double[],double[])
    
    Throws:
    
    IllegalArgumentException - when the length of the two vectors are not the same.
  - averageCommonFeatureRank
```
public static double averageCommonFeatureRank(int[] a,
                              int[] b)
```
    Computes the Average Common Feature Rank between the two feature arrays. Uses the top 20 features for comparison. Converts types and calls averageCommonFeatureRank(Vector,Vector)
    
    Throws:
    
    IllegalArgumentException - when the length of the two vectors are not the same.
  - linSimilarity
```
public static double linSimilarity(DoubleVector a,
                   DoubleVector b)
```
    Computes the lin similarity measure, which is motivated by information theory priniciples. This works best if both vectors have already been weighted using point-wise mutual information. This similarity measure is described in more detail in the following paper:
  - D. Lin, "Automatic Retrieval and Clustering of Similar Words" Proceedings of the 36th Annual Meeting of the Association for Computational Linguistics and 17th International Conference on Computational Linguistics, Volume 2 , Montreal, Quebec, Canada, 1998.

Throws:: IllegalArgumentException - when the length of the two vectors are not the same.

linSimilarity
```
public static double linSimilarity(IntegerVector a,
                   IntegerVector b)
```
Computes the lin similarity measure, which is motivated by information theory priniciples. This works best if both vectors have already been weighted using point-wise mutual information. This similarity measure is described in more detail in the following paper:
D. Lin, "Automatic Retrieval and Clustering of Similar Words" Proceedings of the 36th Annual Meeting of the Association for Computational Linguistics and 17th International Conference on Computational Linguistics, Volume 2 , Montreal, Quebec, Canada, 1998.

Class Similarity

Nested Class Summary

Method Summary

Methods inherited from class java.lang.Object

Method Detail

getMethod

getSimilarityFunction

getSimilarity

getSimilarity

cosineSimilarity

cosineSimilarity

cosineSimilarity

cosineSimilarity

cosineSimilarity

correlation

correlation

correlation

correlation

correlation

euclideanDistance

euclideanDistance

euclideanDistance

euclideanDistance

euclideanDistance

euclideanSimilarity

euclideanSimilarity

euclideanSimilarity

jaccardIndex

jaccardIndex

jaccardIndex

jaccardIndex

jaccardIndex

jaccardIndex

spearmanRankCorrelationCoefficient

spearmanRankCorrelationCoefficient

spearmanRankCorrelationCoefficient

spearmanRankCorrelationCoefficient

spearmanRankCorrelationCoefficient

averageCommonFeatureRank

averageCommonFeatureRank

averageCommonFeatureRank

linSimilarity

linSimilarity

linSimilarity

linSimilarity

linSimilarity

klDivergence

klDivergence

klDivergence

klDivergence

klDivergence

kendallsTau

kendallsTau

kendallsTau

kendallsTau

kendallsTau

goodmanKruskalGamma

tanimotoCoefficient

tanimotoCoefficient

tanimotoCoefficient

tanimotoCoefficient

tanimotoCoefficient