de.aitools.iv.graphdrawing.datastructures.graph.algorithms
Class KNearestNeighbor

java.lang.Object
  de.aitools.iv.graphdrawing.datastructures.graph.algorithms.KNearestNeighbor

All Implemented Interfaces:

GraphProcedure

public class KNearestNeighbor
extends java.lang.Object
implements GraphProcedure

implements an algorithm to compute the k nearest neighbor graph of a weighted graph. NOTE: if the weight of the k-th edge occurs several times in the list of a node's incident edges (a tie), all these occurrences are treated as to be retained. TODO: Introduce tie-handling as an option.

Author:

tim.gollub@medien.uni-weimar.de

Field Summary

static int	MUTUAL
static int	NORMAL

Constructor Summary

KNearestNeighbor()

KNearestNeighbor(int k, int mode)

Method Summary

void	computeKNNGraph(WeightedGraph graph, int k)
void	computeMutualKNNGraph(WeightedGraph graph, int k)
java.lang.Object	execute(WeightedGraph g)
void	setProperties(int k, int mode)

Methods inherited from class java.lang.Object

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

Field Detail

MUTUAL

public static final int MUTUAL

See Also:

Constant Field Values

NORMAL

public static final int NORMAL

See Also:

Constant Field Values

Constructor Detail

KNearestNeighbor

public KNearestNeighbor()

KNearestNeighbor

public KNearestNeighbor(int k,
                        int mode)

Method Detail

setProperties

public void setProperties(int k,
                          int mode)

computeKNNGraph

public void computeKNNGraph(WeightedGraph graph,
                            int k)

computeMutualKNNGraph

public void computeMutualKNNGraph(WeightedGraph graph,
                                  int k)

execute

public java.lang.Object execute(WeightedGraph g)

Specified by:

execute in interface GraphProcedure