Package elki.outlier.lof

Class OnlineLOF<O>

java.lang.Object

elki.outlier.lof.FlexibleLOF<O>

elki.outlier.lof.OnlineLOF<O>

All Implemented Interfaces:

elki.Algorithm, OutlierAlgorithm

public class OnlineLOF<O>
extends FlexibleLOF<O>

Incremental version of the LOF Algorithm, supports insertions and removals.

Since:

0.4.0

Author:

Elke Achtert

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
private class	OnlineLOF.LOFKNNListener	Encapsulates a listener for changes of kNNs used in the online LOF algorithm.
static class	OnlineLOF.Par<O>	Parameterization class.

Nested classes/interfaces inherited from class elki.outlier.lof.FlexibleLOF

FlexibleLOF.LOFResult<O>

Nested classes/interfaces inherited from interface elki.Algorithm

elki.Algorithm.Utils

Field Summary

Fields

Modifier and Type	Field	Description
private static elki.logging.Logging	LOG	The logger for this class.

Fields inherited from class elki.outlier.lof.FlexibleLOF

kreach, krefer, reachabilityDistance, referenceDistance

Constructor Summary

Constructors

Constructor	Description
OnlineLOF(int krefer, int kreach, elki.distance.Distance<? super O> neighborhoodDistance, elki.distance.Distance<? super O> reachabilityDistance)	Constructor.

Method Summary

Modifier and Type	Method	Description
private elki.utilities.pairs.Pair<elki.utilities.pairs.Pair<elki.database.query.knn.KNNSearcher<elki.database.ids.DBIDRef>,elki.database.query.knn.KNNSearcher<elki.database.ids.DBIDRef>>,elki.utilities.pairs.Pair<elki.database.query.rknn.RKNNSearcher<elki.database.ids.DBIDRef>,elki.database.query.rknn.RKNNSearcher<elki.database.ids.DBIDRef>>>	getKNNAndRkNNQueries(elki.database.relation.Relation<O> relation, elki.logging.progress.StepProgress stepprog)	Get the kNN and rkNN queries for the algorithm.
OutlierResult	run(elki.database.relation.Relation<O> relation)	Performs the Generalized LOF_SCORE algorithm on the given database by calling #doRunInTime(Database) and adds a OnlineLOF.LOFKNNListener to the preprocessors.

Methods inherited from class elki.outlier.lof.FlexibleLOF

computeLOFs, computeLRDs, doRunInTime, getInputTypeRestriction

Methods inherited from class java.lang.Object

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

Methods inherited from interface elki.outlier.OutlierAlgorithm

autorun

Field Detail

LOG

private static final elki.logging.Logging LOG

The logger for this class.

Constructor Detail

OnlineLOF

public OnlineLOF(int krefer,
                 int kreach,
                 elki.distance.Distance<? super O> neighborhoodDistance,
                 elki.distance.Distance<? super O> reachabilityDistance)

Constructor.

Parameters:

krefer - The number of neighbors for reference

kreach - The number of neighbors for reachability distance

neighborhoodDistance - the neighborhood distance function

reachabilityDistance - the reachability distance function

Method Detail

run

public OutlierResult run(elki.database.relation.Relation<O> relation)

Performs the Generalized LOF_SCORE algorithm on the given database by calling #doRunInTime(Database) and adds a OnlineLOF.LOFKNNListener to the preprocessors.

Overrides:

run in class FlexibleLOF<O>

Parameters:

relation - Data to process

Returns:

LOF outlier result

getKNNAndRkNNQueries

private elki.utilities.pairs.Pair<elki.utilities.pairs.Pair<elki.database.query.knn.KNNSearcher<elki.database.ids.DBIDRef>,elki.database.query.knn.KNNSearcher<elki.database.ids.DBIDRef>>,elki.utilities.pairs.Pair<elki.database.query.rknn.RKNNSearcher<elki.database.ids.DBIDRef>,elki.database.query.rknn.RKNNSearcher<elki.database.ids.DBIDRef>>> getKNNAndRkNNQueries(elki.database.relation.Relation<O> relation,
                                                                                                                                                                                                                                                                                                                                                                                                elki.logging.progress.StepProgress stepprog)

Get the kNN and rkNN queries for the algorithm.

Parameters:

relation - Data

stepprog - Progress logger

Returns:

the kNN and rkNN queries