de.lmu.ifi.dbs.elki.math.statistics.dependence

Class AbstractDependenceMeasure

java.lang.Object

de.lmu.ifi.dbs.elki.math.statistics.dependence.AbstractDependenceMeasure

All Implemented Interfaces:

DependenceMeasure

Direct Known Subclasses:

CorrelationDependenceMeasure, DistanceCorrelationDependenceMeasure, HiCSDependenceMeasure, HoeffdingsDDependenceMeasure, HSMDependenceMeasure, JensenShannonEquiwidthDependenceMeasure, MCEDependenceMeasure, MutualInformationEquiwidthDependenceMeasure, SlopeDependenceMeasure, SpearmanCorrelationDependenceMeasure, SURFINGDependenceMeasure

public abstract class AbstractDependenceMeasure
extends Object
implements DependenceMeasure

Abstract base class for dependence measures.

Author:

Erich Schubert

Constructor Summary

Constructors

Constructor and Description
AbstractDependenceMeasure()

Method Summary

Methods

Modifier and Type	Method and Description
protected static double	clamp(double value, double min, double max) Clamp values to a given minimum and maximum.
static <A> double[]	computeNormalizedRanks(NumberArrayAdapter<?,A> adapter, A data, int len) Compute ranks of all objects, normalized to [0;1] (where 0 is the smallest value, 1 is the largest).
double	dependence(double[] data1, double[] data2) Measure the dependence of two variables.
<A> double	dependence(NumberArrayAdapter<?,A> adapter, A data1, A data2) Measure the dependence of two variables.
abstract <A,B> double	dependence(NumberArrayAdapter<?,A> adapter1, A data1, NumberArrayAdapter<?,B> adapter2, B data2) Measure the dependence of two variables.
<A> double[]	dependence(NumberArrayAdapter<?,A> adapter, List<? extends A> data) Measure the dependence of two variables.
static <A> int[]	discretize(NumberArrayAdapter<?,A> adapter, A data, int len, int bins) Discretize a data set into equi-width bin numbers.
protected static int	index(int x, int y) Index into the serialized array.
static <A> double[]	ranks(NumberArrayAdapter<?,A> adapter, A data, int len) Compute ranks of all objects, ranging from 1 to len.
static <A> double[]	ranks(NumberArrayAdapter<?,A> adapter, A data, int[] idx) Compute ranks of all objects, ranging from 1 to len.
static <A,B> int	size(NumberArrayAdapter<?,A> adapter1, A data1, NumberArrayAdapter<?,B> adapter2, B data2) Validate the length of the two data sets (must be the same, and non-zero)
static <A> int	size(NumberArrayAdapter<?,A> adapter, Collection<? extends A> data) Validate the length of the two data sets (must be the same, and non-zero)
static <A> int[]	sortedIndex(NumberArrayAdapter<?,A> adapter, A data, int len) Build a sorted index of objects.

Methods inherited from class java.lang.Object

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

Constructor Detail

AbstractDependenceMeasure

public AbstractDependenceMeasure()

Method Detail

dependence

public abstract <A,B> double dependence(NumberArrayAdapter<?,A> adapter1,
                      A data1,
                      NumberArrayAdapter<?,B> adapter2,
                      B data2)

Description copied from interface: DependenceMeasure

Measure the dependence of two variables. This is the more flexible API, which allows using different internal data representations.

Specified by:

dependence in interface DependenceMeasure

Type Parameters:

A - First array type

B - Second array type

Parameters:

adapter1 - First data adapter

data1 - First data set

adapter2 - Second data adapter

data2 - Second data set

Returns:

Dependence measure

dependence

public double dependence(double[] data1,
                double[] data2)

Description copied from interface: DependenceMeasure

Measure the dependence of two variables.

Specified by:

dependence in interface DependenceMeasure

Parameters:

data1 - First data set

data2 - Second data set

Returns:

Dependence measure

dependence

public <A> double dependence(NumberArrayAdapter<?,A> adapter,
                    A data1,
                    A data2)

Description copied from interface: DependenceMeasure

Measure the dependence of two variables. This is the more flexible API, which allows using different internal data representations.

Specified by:

dependence in interface DependenceMeasure

Type Parameters:

A - Array type

Parameters:

adapter - Array type adapter

data1 - First data set

data2 - Second data set

Returns:

Dependence measure

dependence

public <A> double[] dependence(NumberArrayAdapter<?,A> adapter,
                      List<? extends A> data)

Description copied from interface: DependenceMeasure

Measure the dependence of two variables. This is the more flexible API, which allows using different internal data representations. The resulting data is a serialized lower triangular matrix:

  X  S  S  S  S  S
  0  X  S  S  S  S
  1  2  X  S  S  S
  3  4  5  X  S  S
  6  7  8  9  X  S
 10 11 12 13 14  X
 

Specified by:

dependence in interface DependenceMeasure

Type Parameters:

A - Array type

Parameters:

adapter - Data adapter

data - Data sets. Must have fast random access!

Returns:

Lower triangular serialized matrix

clamp

protected static double clamp(double value,
           double min,
           double max)

Clamp values to a given minimum and maximum.

Parameters:

value - True value

min - Minimum

max - Maximum

Returns:

value, unless smaller than min or larger than max.

computeNormalizedRanks

public static <A> double[] computeNormalizedRanks(NumberArrayAdapter<?,A> adapter,
                                  A data,
                                  int len)

Compute ranks of all objects, normalized to [0;1] (where 0 is the smallest value, 1 is the largest).

Parameters:

adapter - Data adapter

data - Data array

len - Length of data

Returns:

Array of scores

ranks

public static <A> double[] ranks(NumberArrayAdapter<?,A> adapter,
                 A data,
                 int len)

Compute ranks of all objects, ranging from 1 to len. Ties are given the average rank.

Parameters:

adapter - Data adapter

data - Data array

len - Length of data

Returns:

Array of scores

ranks

public static <A> double[] ranks(NumberArrayAdapter<?,A> adapter,
                 A data,
                 int[] idx)

Compute ranks of all objects, ranging from 1 to len. Ties are given the average rank.

Parameters:

adapter - Data adapter

data - Data array

idx - Data index

Returns:

Array of scores

sortedIndex

public static <A> int[] sortedIndex(NumberArrayAdapter<?,A> adapter,
                    A data,
                    int len)

Build a sorted index of objects.

Parameters:

adapter - Data adapter

data - Data array

len - Length of data

Returns:

Sorted index

discretize

public static <A> int[] discretize(NumberArrayAdapter<?,A> adapter,
                   A data,
                   int len,
                   int bins)

Discretize a data set into equi-width bin numbers.

Parameters:

adapter - Data adapter

data - Data array

len - Length of data

bins - Number of bins

Returns:

Array of bin numbers [0;bin[

index

protected static int index(int x,
        int y)

Index into the serialized array.

Parameters:

x - Column

y - Row

Returns:

Index in serialized array

size

public static <A,B> int size(NumberArrayAdapter<?,A> adapter1,
             A data1,
             NumberArrayAdapter<?,B> adapter2,
             B data2)

Validate the length of the two data sets (must be the same, and non-zero)

Type Parameters:

A - First array type

B - Second array type

Parameters:

adapter1 - First data adapter

data1 - First data set

adapter2 - Second data adapter

data2 - Second data set

size

public static <A> int size(NumberArrayAdapter<?,A> adapter,
           Collection<? extends A> data)

Validate the length of the two data sets (must be the same, and non-zero)

Type Parameters:

A - First array type

Parameters:

adapter - Data adapter

data - Data sets