de.lmu.ifi.dbs.elki.algorithm.outlier

Class DBOutlierDetection<O,D extends Distance<D>>

java.lang.Object

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

de.lmu.ifi.dbs.elki.algorithm.AbstractDistanceBasedAlgorithm<O,D,OutlierResult>

de.lmu.ifi.dbs.elki.algorithm.outlier.AbstractDBOutlier<O,D>

de.lmu.ifi.dbs.elki.algorithm.outlier.DBOutlierDetection<O,D>

Type Parameters:

O - the type of DatabaseObjects handled by this Algorithm

D - the type of Distance used by this Algorithm

All Implemented Interfaces:

Algorithm, OutlierAlgorithm, Parameterizable

@Title(value="DBOD: Distance Based Outlier Detection")
@Description(value="If the D-neighborhood of an object contains only very few objects (less than (1-p) percent of the data) this object is flagged as an outlier")
@Reference(authors="E.M. Knorr, R. T. Ng",
           title="Algorithms for Mining Distance-Based Outliers in Large Datasets",
           booktitle="Procs Int. Conf. on Very Large Databases (VLDB\'98), New York, USA, 1998")
public class DBOutlierDetection<O,D extends Distance<D>>
extends AbstractDBOutlier<O,D>

Simple distanced based outlier detection algorithm. User has to specify two parameters An object is flagged as an outlier if at least a fraction p of all data objects has a distance above d from c

Reference: E.M. Knorr, R. T. Ng: Algorithms for Mining Distance-Based Outliers in Large Datasets, In: Procs Int. Conf. on Very Large Databases (VLDB'98), New York, USA, 1998. This paper presents several Distance Based Outlier Detection algorithms. Implemented here is a simple index based algorithm as presented in section 3.1.

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	DBOutlierDetection.Parameterizer<O,D extends Distance<D>> Parameterization class.

Field Summary

Fields

Modifier and Type	Field and Description
private static Logging	LOG The logger for this class.
private double	p Holds the value of P_ID.
static OptionID	P_ID Parameter to specify the minimum fraction of objects that must be outside the D- neighborhood of an outlier

Fields inherited from class de.lmu.ifi.dbs.elki.algorithm.outlier.AbstractDBOutlier

D_ID

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

DISTANCE_FUNCTION_ID

Constructor Summary

Constructors

Constructor and Description
DBOutlierDetection(DistanceFunction<O,D> distanceFunction, D d, double p) Constructor with actual parameters.

Method Summary

Methods

Modifier and Type	Method and Description
protected DoubleDataStore	computeOutlierScores(Database database, Relation<O> relation, D neighborhoodSize) computes an outlier score for each object of the database.
protected Logging	getLogger() Get the (STATIC) logger for this class.

Methods inherited from class de.lmu.ifi.dbs.elki.algorithm.outlier.AbstractDBOutlier

getInputTypeRestriction, run

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

Methods inherited from interface de.lmu.ifi.dbs.elki.algorithm.outlier.OutlierAlgorithm

run

Field Detail

LOG

private static final Logging LOG

The logger for this class.

P_ID

public static final OptionID P_ID

Parameter to specify the minimum fraction of objects that must be outside the D- neighborhood of an outlier

p

private double p

Holds the value of P_ID.

Constructor Detail

DBOutlierDetection

public DBOutlierDetection(DistanceFunction<O,D> distanceFunction,
                  D d,
                  double p)

Constructor with actual parameters.

Parameters:

distanceFunction - distance function parameter

d - distance query radius

p - percentage parameter

Method Detail

computeOutlierScores

protected DoubleDataStore computeOutlierScores(Database database,
                                   Relation<O> relation,
                                   D neighborhoodSize)

Description copied from class: AbstractDBOutlier

computes an outlier score for each object of the database.

Specified by:

computeOutlierScores in class AbstractDBOutlier<O,D extends Distance<D>>

Parameters:

database - Database

relation - Relation

neighborhoodSize - distance

Returns:

computed scores

getLogger

protected Logging getLogger()

Description copied from class: AbstractAlgorithm

Get the (STATIC) logger for this class.

Specified by:

getLogger in class AbstractAlgorithm<OutlierResult>

Returns:

the static logger