Package net.sf.tweety.graphs.util

Class GraphUtil

java.lang.Object

net.sf.tweety.graphs.util.GraphUtil

public abstract class GraphUtil
extends Object

This abstract class contains some auxiliary methods for working with graphs.

Author:

Matthias Thimm

Constructor Summary

Constructors

Constructor	Description
GraphUtil()

Method Summary

Modifier and Type	Method	Description
static <T extends Node> Map<T,Double>	betweennessCentralityNormalised(Graph<T> graph)	Computes the normalised betweenness centrality of all nodes, i.e.
static net.sf.tweety.math.ComplexNumber[]	eigenvalues(Graph<? extends Node> g)	Computes the (real parts of the) Eigenvalues of the given graph.
static <T extends Node> Collection<List<T>>	enumerateChordlessCircuits(Graph<T> g)	Enumerates all chordless circuits of the given graph, i.e.
static <T extends Node> double	globalclusteringcoefficient(Graph<T> g)	Returns the global clustering coefficient of the graph (if it is directed it is interpreted as an undirected version).
static Double	hitsRank(Graph<? extends Node> g, Node n, double precision, boolean getAuth)	Computes the HITS rank of the given node in the given graph.
static boolean	isIsomorphic(Graph<? extends Node> g1, Graph<? extends Node> g2)	Checks whether the two graphs are isomorphic.
static Double	pageRank(Graph<? extends Node> g, Node n, double dampingFactor, double precision)	Computes the PageRank of the given node in the given graph.
static <T extends Node> int	undirecteddiameter(Graph<T> g)	Returns the (undirected) diameter of the graph, i.e.

Methods inherited from class java.lang.Object

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

Constructor Detail

GraphUtil

public GraphUtil()

Method Detail

pageRank

public static Double pageRank(Graph<? extends Node> g,
                              Node n,
                              double dampingFactor,
                              double precision)

Computes the PageRank of the given node in the given graph.

Parameters:

g - a graph

n - a node

dampingFactor - the damping factor for PageRank

precision - the precision (smaller values mean higher precision)

Returns:

the PageRank of the given node in the given graph.

hitsRank

public static Double hitsRank(Graph<? extends Node> g,
                              Node n,
                              double precision,
                              boolean getAuth)

Computes the HITS rank of the given node in the given graph.

Parameters:

g - a graph

n - a node

precision - the precision (smaller values mean higher precision)

getAuth - whether to use Auth (instead of Hub)

Returns:

the HITS rank of the given node in the given graph.

eigenvalues

public static net.sf.tweety.math.ComplexNumber[] eigenvalues(Graph<? extends Node> g)

Computes the (real parts of the) Eigenvalues of the given graph.

Parameters:

g - some graph

Returns:

an array of double (the real parts of the Eigenvalues).

isIsomorphic

public static boolean isIsomorphic(Graph<? extends Node> g1,
                                   Graph<? extends Node> g2)

Checks whether the two graphs are isomorphic.

Parameters:

g1 - some graph.

g2 - some graph.

Returns:

"true" iff the two graphs are isomorphic.

undirecteddiameter

public static <T extends Node> int undirecteddiameter(Graph<T> g)

Returns the (undirected) diameter of the graph, i.e. the longest shortest path in the undirected version of the graph.

Parameters:

g - some graph

Returns:

the (undirected) diameter of the graph

globalclusteringcoefficient

public static <T extends Node> double globalclusteringcoefficient(Graph<T> g)

Returns the global clustering coefficient of the graph (if it is directed it is interpreted as an undirected version).

Parameters:

g - some graph

Returns:

the clustering coefficient

enumerateChordlessCircuits

public static <T extends Node> Collection<List<T>> enumerateChordlessCircuits(Graph<T> g)

Enumerates all chordless circuits of the given graph, i.e. all circuits a1,...,an where there is no edge connecting any ak with aj unless k=j+1 or k=j-1. The algorithm of this method is adapted from [Bisdorff, On enumerating chordless circuits in directed graphs, 2010].

Parameters:

g - some graph

Returns:

the set of chordless circuits

betweennessCentralityNormalised

public static <T extends Node> Map<T,Double> betweennessCentralityNormalised(Graph<T> graph)

Computes the normalised betweenness centrality of all nodes, i.e. the number of shortest paths going through each node. The value is normalised by subtracting the minimum number (min) of such shortest paths and dividing by (max-min).

Parameters:

graph - some graph

Returns:

a map mapping each node to its betweenness centrality.