de.lmu.ifi.dbs.elki.algorithm.outlier
Class LoOP<O extends DatabaseObject>

java.lang.Object
  de.lmu.ifi.dbs.elki.logging.AbstractLoggable
      de.lmu.ifi.dbs.elki.algorithm.AbstractAlgorithm<O,MultiResult>
          de.lmu.ifi.dbs.elki.algorithm.outlier.LoOP<O>

Type Parameters:

O - the type of DatabaseObjects handled by this Algorithm

All Implemented Interfaces:

Algorithm<O,MultiResult>, Parameterizable

@Title(value="LoOP: Local Outlier Probabilities")
@Description(value="Variant of the LOF algorithm normalized using statistical values.")
@Reference(authors="H.-P. Kriegel, P. Kr\u00f6ger, E. Schubert, A. Zimek",
           title="LoOP: Local Outlier Probabilities",
           booktitle="Proceedings of the 18th International Conference on Information and Knowledge Management (CIKM), Hong Kong, China, 2009",
           url="http://dx.doi.org/10.1145/1645953.1646195")
public class LoOP<O extends DatabaseObject>
extends AbstractAlgorithm<O,MultiResult>

LoOP: Local Outlier Probabilities Distance/density based algorithm similar to LOF to detect outliers, but with statistical methods to achieve better result stability.

Author:

Erich Schubert

Field Summary

static OptionID	COMPARISON_DISTANCE_FUNCTION_ID OptionID for COMPARISON_DISTANCE_FUNCTION_PARAM
private ObjectParameter<DistanceFunction<O,DoubleDistance>>	COMPARISON_DISTANCE_FUNCTION_PARAM The distance function to determine the reachability distance between database objects.
(package private) int	kcomp Holds the value of KCOMP_PARAM.
static OptionID	KCOMP_ID OptionID for KCOMP_PARAM
private IntParameter	KCOMP_PARAM Parameter to specify the number of nearest neighbors of an object to be considered for computing its LOOP_SCORE, must be an integer greater than 1.
(package private) int	kref Holds the value of KREF_PARAM.
static OptionID	KREF_ID OptionID for KCOMP_PARAM
private IntParameter	KREF_PARAM Parameter to specify the number of nearest neighbors of an object to be considered for computing its LOOP_SCORE, must be an integer greater than 1.
(package private) double	lambda Hold the value of LAMBDA_PARAM.
static OptionID	LAMBDA_ID OptionID for LAMBDA_PARAM
private DoubleParameter	LAMBDA_PARAM Parameter to specify the number of nearest neighbors of an object to be considered for computing its LOOP_SCORE, must be an integer greater than 1.
static AssociationID<Double>	LOOP_SCORE The association id to associate the LOOP_SCORE of an object for the LOOP_SCORE algorithm.
(package private) boolean	objectIsInKNN Include object itself in kNN neighborhood.
static OptionID	PREPROCESSOR_ID OptionID for PREPROCESSOR_PARAM
private ClassParameter<MaterializeKNNPreprocessor<O,DoubleDistance>>	PREPROCESSOR_PARAM The preprocessor used to materialize the kNN neighborhoods.
(package private) MaterializeKNNPreprocessor<O,DoubleDistance>	preprocessorcompare Preprocessor Step 1
(package private) MaterializeKNNPreprocessor<O,DoubleDistance>	preprocessorref Preprocessor Step 2
static OptionID	REFERENCE_DISTANCE_FUNCTION_ID OptionID for REFERENCE_DISTANCE_FUNCTION_PARAM
private ObjectParameter<DistanceFunction<O,DoubleDistance>>	REFERENCE_DISTANCE_FUNCTION_PARAM The distance function to determine the reachability distance between database objects.

Fields inherited from class de.lmu.ifi.dbs.elki.logging.AbstractLoggable

debug, logger

Constructor Summary

LoOP(Parameterization config)
Constructor, adhering to Parameterizable

Method Summary

protected OutlierResult

runInTime(Database<O> database) Performs the LoOP algorithm on the given database.

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

isTime, isVerbose, run, setTime, setVerbose

Methods inherited from class de.lmu.ifi.dbs.elki.logging.AbstractLoggable

debugFine, debugFiner, debugFinest, exception, progress, verbose, warning

Methods inherited from class java.lang.Object

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

Field Detail

REFERENCE_DISTANCE_FUNCTION_ID

public static final OptionID REFERENCE_DISTANCE_FUNCTION_ID

OptionID for REFERENCE_DISTANCE_FUNCTION_PARAM

REFERENCE_DISTANCE_FUNCTION_PARAM

private final ObjectParameter<DistanceFunction<O extends DatabaseObject,DoubleDistance>> REFERENCE_DISTANCE_FUNCTION_PARAM

The distance function to determine the reachability distance between database objects.

Default value: EuclideanDistanceFunction

Key: -loop.referencedistfunction

COMPARISON_DISTANCE_FUNCTION_ID

public static final OptionID COMPARISON_DISTANCE_FUNCTION_ID

OptionID for COMPARISON_DISTANCE_FUNCTION_PARAM

COMPARISON_DISTANCE_FUNCTION_PARAM

private final ObjectParameter<DistanceFunction<O extends DatabaseObject,DoubleDistance>> COMPARISON_DISTANCE_FUNCTION_PARAM

The distance function to determine the reachability distance between database objects.

Default value: EuclideanDistanceFunction

Key: -loop.comparedistfunction

PREPROCESSOR_ID

public static final OptionID PREPROCESSOR_ID

OptionID for PREPROCESSOR_PARAM

PREPROCESSOR_PARAM

private final ClassParameter<MaterializeKNNPreprocessor<O extends DatabaseObject,DoubleDistance>> PREPROCESSOR_PARAM

The preprocessor used to materialize the kNN neighborhoods. Default value: MaterializeKNNPreprocessor

Key: -loop.preprocessor

LOOP_SCORE

public static final AssociationID<Double> LOOP_SCORE

The association id to associate the LOOP_SCORE of an object for the LOOP_SCORE algorithm.

KCOMP_ID

public static final OptionID KCOMP_ID

OptionID for KCOMP_PARAM

KCOMP_PARAM

private final IntParameter KCOMP_PARAM

Parameter to specify the number of nearest neighbors of an object to be considered for computing its LOOP_SCORE, must be an integer greater than 1.

Key: -loop.kcomp

KREF_ID

public static final OptionID KREF_ID

OptionID for KCOMP_PARAM

KREF_PARAM

private final IntParameter KREF_PARAM

Parameter to specify the number of nearest neighbors of an object to be considered for computing its LOOP_SCORE, must be an integer greater than 1.

Key: -loop.kref

LAMBDA_ID

public static final OptionID LAMBDA_ID

OptionID for LAMBDA_PARAM

LAMBDA_PARAM

private final DoubleParameter LAMBDA_PARAM

Parameter to specify the number of nearest neighbors of an object to be considered for computing its LOOP_SCORE, must be an integer greater than 1.

Key: -loop.lambda

kcomp

int kcomp

Holds the value of KCOMP_PARAM.

kref

int kref

Holds the value of KREF_PARAM.

lambda

double lambda

Hold the value of LAMBDA_PARAM.

preprocessorcompare

MaterializeKNNPreprocessor<O extends DatabaseObject,DoubleDistance> preprocessorcompare

Preprocessor Step 1

preprocessorref

MaterializeKNNPreprocessor<O extends DatabaseObject,DoubleDistance> preprocessorref

Preprocessor Step 2

objectIsInKNN

boolean objectIsInKNN

Include object itself in kNN neighborhood.

Constructor Detail

LoOP

public LoOP(Parameterization config)

Constructor, adhering to Parameterizable

Parameters:

config - Parameterization

Method Detail

runInTime

protected OutlierResult runInTime(Database<O> database)
                           throws IllegalStateException

Performs the LoOP algorithm on the given database.

Specified by:

runInTime in class AbstractAlgorithm<O extends DatabaseObject,MultiResult>

Parameters:

database - the database to run the algorithm on

Returns:

the Result computed by this algorithm

Throws:

IllegalStateException - if the algorithm has not been initialized properly (e.g. the setParameters(String[]) method has been failed to be called).