Package elki.outlier.meta

Class FeatureBagging

java.lang.Object

elki.outlier.meta.FeatureBagging

All Implemented Interfaces:

Algorithm, OutlierAlgorithm

@Title("Feature Bagging for Outlier Detection")
@Reference(authors="A. Lazarevic, V. Kumar",
           title="Feature Bagging for Outlier Detection",
           booktitle="Proc. 11th ACM SIGKDD Int. Conf. on Knowledge Discovery in Data Mining",
           url="https://doi.org/10.1145/1081870.1081891",
           bibkey="DBLP:conf/kdd/LazarevicK05")
public class FeatureBagging
extends java.lang.Object
implements OutlierAlgorithm

A simple ensemble method called "Feature bagging" for outlier detection.

Since the proposed method is only sensible to run on multiple instances of the same algorithm (due to incompatible score ranges), we do not allow using arbitrary algorithms.

Reference:
A. Lazarevic, V. Kumar
Feature Bagging for Outlier Detection
Proc. 11th ACM SIGKDD Int. Conf. on Knowledge Discovery in Data Mining

Since:

0.4.0

Author:

Erich Schubert, Remigius Wojdanowski

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
static class	FeatureBagging.Par	Parameterization class.

Nested classes/interfaces inherited from interface elki.Algorithm

Algorithm.Utils

Field Summary

Fields

Modifier and Type	Field	Description
protected boolean	breadth	Cumulative sum or breadth first combinations.
private int	k	The parameters k for LOF.
private static Logging	LOG	The logger for this class.
protected int	num	Number of instances to use.
private RandomFactory	rnd	Random number generator for subspace choice.

Constructor Summary

Constructors

Constructor	Description
FeatureBagging(int k, int num, boolean breadth, RandomFactory rnd)	Constructor.

Method Summary

Modifier and Type	Method	Description
TypeInformation[]	getInputTypeRestriction()	Get the input type restriction used for negotiating the data query.
private long[]	randomSubspace(int alldim, int mindim, int maxdim, java.util.Random rand)	Choose a random subspace.
OutlierResult	run(Relation<NumberVector> relation)	Run the algorithm on a data set.

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 Logging LOG

The logger for this class.

num

protected int num

Number of instances to use.

breadth

protected boolean breadth

Cumulative sum or breadth first combinations.

rnd

private RandomFactory rnd

Random number generator for subspace choice.

k

private int k

The parameters k for LOF.

Constructor Detail

FeatureBagging

public FeatureBagging(int k,
                      int num,
                      boolean breadth,
                      RandomFactory rnd)

Constructor.

Parameters:

k - k Parameter for LOF

num - Number of subspaces to use

breadth - Flag for breadth-first merging

rnd - Random generator

Method Detail

getInputTypeRestriction

public TypeInformation[] getInputTypeRestriction()

Description copied from interface: Algorithm

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

Specified by:

getInputTypeRestriction in interface Algorithm

Returns:

Type restriction

run

public OutlierResult run(Relation<NumberVector> relation)

Run the algorithm on a data set.

Parameters:

relation - Relation to use

Returns:

Outlier detection result

randomSubspace

private long[] randomSubspace(int alldim,
                              int mindim,
                              int maxdim,
                              java.util.Random rand)

Choose a random subspace.

Parameters:

alldim - Number of total dimensions

mindim - Minimum number to choose

maxdim - Maximum number to choose

Returns:

Subspace as bits.