jsat.linear.distancemetrics

Class PearsonDistance

java.lang.Object

jsat.linear.distancemetrics.PearsonDistance

All Implemented Interfaces:

java.io.Serializable, java.lang.Cloneable, DistanceMetric

public class PearsonDistance
extends java.lang.Object
implements DistanceMetric

A valid distance metric formed from the Pearson Correlation between two vectors. The distance in the range of [0, 1].

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
PearsonDistance() Creates a new standard Pearson Distance that does not ignore zero values and anti-correlated values are considered far away.
PearsonDistance(boolean bothNonZero, boolean absoluteDistance) Creates a new Pearson Distance object

Method Summary

Modifier and Type	Method and Description
PearsonDistance	clone()
static double	correlation(Vec a, Vec b, boolean bothNonZero) Computes the Pearson correlation between two vectors.
double	dist(Vec a, Vec b) Computes the distance between 2 vectors.
boolean	isIndiscemible() Returns true if this distance metric obeys the rule that, for any x and y ∈ S d(x, y) = 0 if and only if x = y
boolean	isSubadditive() Returns true if this distance metric obeys the rule that, for any x, y, and z ∈ S d(x, z) ≤ d(x, y) + d(y, z)
boolean	isSymmetric() Returns true if this distance metric obeys the rule that, for any x, y, and z ∈ S d(x, y) = d(y, x)
double	metricBound() All metrics must return values greater than or equal to 0.

Methods inherited from class java.lang.Object

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

Methods inherited from interface jsat.linear.distancemetrics.DistanceMetric

dist, dist, dist, getAccelerationCache, getAccelerationCache, getQueryInfo, isValidMetric, supportsAcceleration, toString

Constructor Detail

PearsonDistance

public PearsonDistance()

Creates a new standard Pearson Distance that does not ignore zero values and anti-correlated values are considered far away.

PearsonDistance

public PearsonDistance(boolean bothNonZero,
                       boolean absoluteDistance)

Creates a new Pearson Distance object

Parameters:

bothNonZero - true if non zero values should be treated as "missing" or "no vote", and will not contribute. But this will not change the mean value used. false produces the standard Pearson value.

absoluteDistance - true to use the absolute correlation, meaning correlated and anti-correlated values will have the same distance.

Method Detail

dist

public double dist(Vec a,
                   Vec b)

Description copied from interface: DistanceMetric

Computes the distance between 2 vectors. The smaller the value, the closer, and there for, more similar, the vectors are. 0 indicates the vectors are the same.

Specified by:

dist in interface DistanceMetric

Parameters:

a - the first vector

b - the second vector

Returns:

the distance between them

isSymmetric

public boolean isSymmetric()

Description copied from interface: DistanceMetric

Returns true if this distance metric obeys the rule that, for any x, y, and z ∈ S
d(x, y) = d(y, x)

Specified by:

isSymmetric in interface DistanceMetric

Returns:

true if this distance metric is symmetric, false if it is not

isSubadditive

public boolean isSubadditive()

Description copied from interface: DistanceMetric

Returns true if this distance metric obeys the rule that, for any x, y, and z ∈ S
d(x, z) ≤ d(x, y) + d(y, z)

Specified by:

isSubadditive in interface DistanceMetric

Returns:

true if this distance metric supports the triangle inequality, false if it does not.

isIndiscemible

public boolean isIndiscemible()

Description copied from interface: DistanceMetric

Returns true if this distance metric obeys the rule that, for any x and y ∈ S
d(x, y) = 0 if and only if x = y

Specified by:

isIndiscemible in interface DistanceMetric

Returns:

true if this distance metric is indicemible, false otherwise.

metricBound

public double metricBound()

Description copied from interface: DistanceMetric

All metrics must return values greater than or equal to 0. The upper bound on the value returned is different for different metrics. This method returns the theoretical maximal value that could be returned by this distance metric. That means Double.POSITIVE_INFINITY is a valid return value.

Specified by:

metricBound in interface DistanceMetric

Returns:

the maximal distance for any two points in that could exist by this distance metric.

clone

public PearsonDistance clone()

Specified by:

clone in interface DistanceMetric

Overrides:

clone in class java.lang.Object

correlation

public static double correlation(Vec a,
                                 Vec b,
                                 boolean bothNonZero)

Computes the Pearson correlation between two vectors. If one of the vectors is all zeros, the result is undefined. In cases where both are zero vectors, 1 will be returned to indicate they are the same. In cases where one of the numerator coefficients is zero, its value will be bumped up to an epsilon to provide a near result.

In cases where bothNonZero is true, and the vectors have no overlapping non zero values, 0 will be returned.

Parameters:

a - the first vector

b - the second vector

bothNonZero - false is the normal Pearson correlation. true will make the computation ignore all indexes where one of the values is zero, the mean will be from all non zero values in each vector.

Returns:

the Pearson correlation in [-1, 1]