Package com.gengoai.graph.clustering

Class ChineseWhispers<V>

java.lang.Object

com.gengoai.graph.clustering.ChineseWhispers<V>

Type Parameters:

V - the type parameter

All Implemented Interfaces:

Clusterer<V>

public class ChineseWhispers<V>
extends Object
implements Clusterer<V>

Implementation of the Chinese Whispers graph clustering algorithm as outlined in

• Biemann, Chris. "Chinese whispers: an efficient graph clustering algorithm and its application to natural language processing problems." Proceedings of the First Workshop on Graph Based Methods for Natural Language Processing. Association for Computational Linguistics, 2006.

Author:

David B. Bracewell

Constructor Summary

Constructors

Constructor	Description
ChineseWhispers()	Default constructor maxIterations = 10, acceptanceRate = 1, and newClusterRate = 0d
ChineseWhispers(int maxIterations, double acceptanceRate, double newClusterRate)	Initialization constructor

Method Summary

Modifier and Type	Method	Description
List<Set<V>>	cluster(Graph<V> g)	Clusters the vertices of the given graph and returns a list of clusters (sets).
double	getAcceptanceRate()	Gets acceptance rate.
int	getMaxIterations()	Gets max iterations.
double	getNewClusterRate()	Gets new cluster rate.
void	setAcceptanceRate(double acceptanceRate)	Sets acceptance rate.
void	setMaxIterations(int maxIterations)	Sets max iterations.
void	setNewClusterRate(double newClusterRate)	Sets new cluster rate.

Methods inherited from class java.lang.Object

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

Constructor Detail

ChineseWhispers

public ChineseWhispers()

Default constructor maxIterations = 10, acceptanceRate = 1, and newClusterRate = 0d

ChineseWhispers

public ChineseWhispers(int maxIterations,
                       double acceptanceRate,
                       double newClusterRate)

Initialization constructor

Parameters:

maxIterations - The maximum number of iterations to run the clustering

acceptanceRate - The rate used to determine if a label change should be accepted

newClusterRate - the new cluster rate

Method Detail

getNewClusterRate

public double getNewClusterRate()

Gets new cluster rate.

Returns:

the new cluster rate

setNewClusterRate

public void setNewClusterRate(double newClusterRate)

Sets new cluster rate.

Parameters:

newClusterRate - the new cluster rate

cluster

public List<Set<V>> cluster(Graph<V> g)

Description copied from interface: Clusterer

Clusters the vertices of the given graph and returns a list of clusters (sets).

Specified by:

cluster in interface Clusterer<V>

Parameters:

g - the graph

Returns:

the clusters

getMaxIterations

public int getMaxIterations()

Gets max iterations.

Returns:

the maxIterations

setMaxIterations

public void setMaxIterations(int maxIterations)

Sets max iterations.

Parameters:

maxIterations - the maxIterations to set

getAcceptanceRate

public double getAcceptanceRate()

Gets acceptance rate.

Returns:

the acceptanceRate

setAcceptanceRate

public void setAcceptanceRate(double acceptanceRate)

Sets acceptance rate.

Parameters:

acceptanceRate - the acceptanceRate to set