org.graphstream.algorithm.measure
Class SurpriseMeasure

java.lang.Object
  org.graphstream.algorithm.measure.SurpriseMeasure

All Implemented Interfaces:

Algorithm

public class SurpriseMeasure
extends Object
implements Algorithm

Surprise measure. Description from Wikipedia : Surprise (denoted S) is a measure of community structure in complex networks. The name Surprise derives from the fact that its maximization finds the most surprising partition into communities of the network, that is, the most unlikely one. S accurately evaluates, in a global manner, the quality of a partition using a cumulative hypergeometric distribution.

Scientific Reference :

Rodrigo Aldecoa, Ignacio Marin, "Deciphering Network Community Structure by Surprise", 2011, PLoS ONE 6(9)

Field Summary

static String

ATTRIBUTE Default attribute key where the result of the algorithm, a double value, is stored.

Constructor Summary

SurpriseMeasure()
Default constructor.

SurpriseMeasure(String communityAttributeKey)
Constructor allowing to set the node attribute key containing index of organizations.

SurpriseMeasure(String communityAttributeKey, String surpriseAttributeKey)
Same as SurpriseMeasure(String) but allowing to set the graph attribute that will contain the result of the computation.

Method Summary

static double	binomialCoefficient(double n, double r) Helper to compute the binomial coefficient.
void	compute() Run the algorithm.
static double	cumulativeHypergeometricDistribution(double xStart, double xEnd, double N, double n, double k) Helper to compute the cumulative hypergeometric distribution.
double	getSurprise() Get the last computed surprise value contained in the graph.
static double	hypergeometricDistribution(double x, double N, double n, double k) Helper to compute the hypergeometric distribution.
void	init(org.graphstream.graph.Graph graph) Initialization of the algorithm.

Methods inherited from class java.lang.Object

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

Field Detail

ATTRIBUTE

public static final String ATTRIBUTE

Default attribute key where the result of the algorithm, a double value, is stored.

See Also:

Constant Field Values

Constructor Detail

SurpriseMeasure

public SurpriseMeasure()

Default constructor.

SurpriseMeasure

public SurpriseMeasure(String communityAttributeKey)

Constructor allowing to set the node attribute key containing index of organizations.

Parameters:

communityAttributeKey - key attribute of organizations

SurpriseMeasure

public SurpriseMeasure(String communityAttributeKey,
                       String surpriseAttributeKey)

Same as SurpriseMeasure(String) but allowing to set the graph attribute that will contain the result of the computation.

Parameters:

communityAttributeKey -

surpriseAttributeKey -

Method Detail

init

public void init(org.graphstream.graph.Graph graph)

Description copied from interface: Algorithm

Initialization of the algorithm. This method has to be called before the Algorithm.compute() method to initialize or reset the algorithm according to the new given graph.

Specified by:

init in interface Algorithm

Parameters:

graph - The graph this algorithm is using.

compute

public void compute()

Description copied from interface: Algorithm

Run the algorithm. The Algorithm.init(Graph) method has to be called before computing.

Specified by:

compute in interface Algorithm

See Also:

Algorithm.init(Graph)

getSurprise

public double getSurprise()

Get the last computed surprise value contained in the graph.

Returns:

surprise value

binomialCoefficient

public static double binomialCoefficient(double n,
                                         double r)

Helper to compute the binomial coefficient.

Parameters:

n -

r -

Returns:

hypergeometricDistribution

public static double hypergeometricDistribution(double x,
                                                double N,
                                                double n,
                                                double k)

Helper to compute the hypergeometric distribution. See this page for more information about this function.

Parameters:

x -

N -

n -

k -

Returns:

cumulativeHypergeometricDistribution

public static double cumulativeHypergeometricDistribution(double xStart,
                                                          double xEnd,
                                                          double N,
                                                          double n,
                                                          double k)

Helper to compute the cumulative hypergeometric distribution. See this page for more information about this function.

Parameters:

xStart -

xEnd -

N -

n -

k -

Returns: