de.lmu.ifi.dbs.elki.algorithm

Class KNNJoin<V extends NumberVector<V,?>,D extends Distance<D>,N extends SpatialNode<N,E>,E extends SpatialEntry>

java.lang.Object

de.lmu.ifi.dbs.elki.algorithm.AbstractAlgorithm<R>

de.lmu.ifi.dbs.elki.algorithm.AbstractDistanceBasedAlgorithm<V,D,DataStore<KNNList<D>>>

de.lmu.ifi.dbs.elki.algorithm.KNNJoin<V,D,N,E>

Type Parameters:

V - the type of FeatureVector handled by this Algorithm

D - the type of Distance used by this Algorithm

N - the type of node used in the spatial index structure

E - the type of entry used in the spatial node

All Implemented Interfaces:

Algorithm, InspectionUtilFrequentlyScanned, Parameterizable

@Title(value="K-Nearest Neighbor Join")
@Description(value="Algorithm to find the k-nearest neighbors of each object in a spatial database")
public class KNNJoin<V extends NumberVector<V,?>,D extends Distance<D>,N extends SpatialNode<N,E>,E extends SpatialEntry>
extends AbstractDistanceBasedAlgorithm<V,D,DataStore<KNNList<D>>>

Joins in a given spatial database to each object its k-nearest neighbors. This algorithm only supports spatial databases based on a spatial index structure. Since this method compares the MBR of every single leaf with every other leaf, it is essentially quadratic in the number of leaves, which may not be appropriate for large trees.

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	KNNJoin.Parameterizer<V extends NumberVector<V,?>,D extends Distance<D>,N extends SpatialNode<N,E>,E extends SpatialEntry> Parameterization class.
private class	KNNJoin.Task Task in the processing queue

Field Summary

Fields

Modifier and Type	Field and Description
(package private) int	k The k parameter
static OptionID	K_ID Parameter that specifies the k-nearest neighbors to be assigned, must be an integer greater than 0.
private static Logging	logger The logger for this class.

Fields inherited from class de.lmu.ifi.dbs.elki.algorithm.AbstractDistanceBasedAlgorithm

DISTANCE_FUNCTION_ID

Constructor Summary

Constructors

Constructor and Description
KNNJoin(DistanceFunction<? super V,D> distanceFunction, int k) Constructor.

Method Summary

Methods

Modifier and Type	Method and Description
private D	computeStopDistance(List<KNNHeap<D>> heaps) Compute the maximum stop distance
TypeInformation[]	getInputTypeRestriction() Get the input type restriction used for negotiating the data query.
protected Logging	getLogger() Get the (STATIC) logger for this class.
private List<KNNHeap<D>>	initHeaps(SpatialPrimitiveDistanceFunction<V,D> distFunction, boolean doubleOptimize, N pr)
private void	processDataPagesDouble(SpatialPrimitiveDoubleDistanceFunction<? super V> df, N pr, N ps, List<KNNHeap<DoubleDistance>> pr_heaps, List<KNNHeap<DoubleDistance>> ps_heaps) Processes the two data pages pr and ps and determines the k-nearest neighbors of pr in ps.
private void	processDataPagesOptimize(SpatialPrimitiveDistanceFunction<V,D> distFunction, boolean doubleOptimize, List<KNNHeap<D>> pr_heaps, List<KNNHeap<D>> ps_heaps, N pr, N ps) Processes the two data pages pr and ps and determines the k-nearest neighbors of pr in ps.
WritableDataStore<KNNList<D>>	run(Database database, Relation<V> relation) Joins in the given spatial database to each object its k-nearest neighbors.

Methods inherited from class de.lmu.ifi.dbs.elki.algorithm.AbstractDistanceBasedAlgorithm

getDistanceFunction

Methods inherited from class de.lmu.ifi.dbs.elki.algorithm.AbstractAlgorithm

makeParameterDistanceFunction, run

Methods inherited from class java.lang.Object

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

Field Detail

logger

private static final Logging logger

The logger for this class.

K_ID

public static final OptionID K_ID

Parameter that specifies the k-nearest neighbors to be assigned, must be an integer greater than 0. Default value: 1.

k

int k

The k parameter

Constructor Detail

KNNJoin

public KNNJoin(DistanceFunction<? super V,D> distanceFunction,
       int k)

Constructor.

Parameters:

distanceFunction - Distance function

k - k parameter

Method Detail

run

public WritableDataStore<KNNList<D>> run(Database database,
                                Relation<V> relation)
                                                      throws IllegalStateException

Joins in the given spatial database to each object its k-nearest neighbors.

Throws:

IllegalStateException - if not suitable SpatialIndexTree was found or the specified distance function is not an instance of SpatialPrimitiveDistanceFunction.

initHeaps

private List<KNNHeap<D>> initHeaps(SpatialPrimitiveDistanceFunction<V,D> distFunction,
                         boolean doubleOptimize,
                         N pr)

processDataPagesOptimize

private void processDataPagesOptimize(SpatialPrimitiveDistanceFunction<V,D> distFunction,
                            boolean doubleOptimize,
                            List<KNNHeap<D>> pr_heaps,
                            List<KNNHeap<D>> ps_heaps,
                            N pr,
                            N ps)

Processes the two data pages pr and ps and determines the k-nearest neighbors of pr in ps.

Parameters:

distFunction - the distance to use

doubleOptimize - Flag whether to optimize for doubles.

pr - the first data page

ps - the second data page

pr_heaps - the knn lists for each data object in pr

ps_heaps - the knn lists for each data object in ps (if ps != pr)

processDataPagesDouble

private void processDataPagesDouble(SpatialPrimitiveDoubleDistanceFunction<? super V> df,
                          N pr,
                          N ps,
                          List<KNNHeap<DoubleDistance>> pr_heaps,
                          List<KNNHeap<DoubleDistance>> ps_heaps)

Processes the two data pages pr and ps and determines the k-nearest neighbors of pr in ps.

Parameters:

df - the distance function to use

pr - the first data page

ps - the second data page

pr_heaps - the knn lists for each data object

ps_heaps - the knn lists for each data object in ps

computeStopDistance

private D computeStopDistance(List<KNNHeap<D>> heaps)

Compute the maximum stop distance

Parameters:

heaps -

Returns:

the k-nearest neighbor distance of pr in ps

getInputTypeRestriction

public TypeInformation[] getInputTypeRestriction()

Description copied from class: AbstractAlgorithm

Get the input type restriction used for negotiating the data query.

Specified by:

getInputTypeRestriction in interface Algorithm

Specified by:

getInputTypeRestriction in class AbstractAlgorithm<DataStore<KNNList<D extends Distance<D>>>>

Returns:

Type restriction

getLogger

protected Logging getLogger()

Description copied from class: AbstractAlgorithm

Get the (STATIC) logger for this class.

Specified by:

getLogger in class AbstractAlgorithm<DataStore<KNNList<D extends Distance<D>>>>

Returns:

the static logger