uk.ac.kent.dover.fastGraph

Class KMedoids

java.lang.Object

uk.ac.kent.dover.fastGraph.KMedoids

public class KMedoids
extends java.lang.Object

Algorithm to implement the KMedoids Modified from https://sourceforge.net/p/java-ml/java-ml-code/ci/a25ddde7c3677da44e47a643f88e32e2c8bbc32f/tree/net/sf/javaml/clustering/KMedoids.java#l40

Field Summary

Fields

Modifier and Type	Field and Description
long	gedTime
int	numberOfGedCalcs

Constructor Summary

Constructors

Constructor and Description
KMedoids(FastGraph targetGraph, int numberOfClusters, int maxIterations) Constructor

Method Summary

Modifier and Type	Method and Description
java.util.ArrayList<java.util.ArrayList<FastGraph>>	cluster(java.util.ArrayList<FastGraph> subgraphs) Clusters the subgraphs
double	comparisonScore(FastGraph g1, FastGraph g2) Returns the comparison score of the two graphs.
void	saveClusters(java.util.ArrayList<java.util.ArrayList<FastGraph>> clusters) Save clusters to disk

Methods inherited from class java.lang.Object

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

Field Detail

numberOfGedCalcs

public int numberOfGedCalcs

gedTime

public long gedTime

Constructor Detail

KMedoids

public KMedoids(FastGraph targetGraph,
                int numberOfClusters,
                int maxIterations)

Constructor

Parameters:

targetGraph - The target graph

numberOfClusters - The number of clusters

maxIterations - The maximum number of iterations

Method Detail

cluster

public java.util.ArrayList<java.util.ArrayList<FastGraph>> cluster(java.util.ArrayList<FastGraph> subgraphs)
                                                            throws FastGraphException

Clusters the subgraphs

Parameters:

subgraphs - The subgraphs to cluster

Returns:

The clusters, as a list of lists of FastGraphs

Throws:

FastGraphException - If the random selection cannot be obtained (i.e. k is too big)

comparisonScore

public double comparisonScore(FastGraph g1,
                              FastGraph g2)

Returns the comparison score of the two graphs. Normally, GED

Parameters:

g1 - The first graph

g2 - The second graph

Returns:

The comparison score

saveClusters

public void saveClusters(java.util.ArrayList<java.util.ArrayList<FastGraph>> clusters)
                  throws java.io.IOException

Save clusters to disk

Parameters:

clusters - The clusters to save

Throws:

java.io.IOException