de.lmu.ifi.dbs.elki.application.greedyensemble

Class ComputeKNNOutlierScores<O,D extends NumberDistance<D,?>>

java.lang.Object

de.lmu.ifi.dbs.elki.application.AbstractApplication

de.lmu.ifi.dbs.elki.application.greedyensemble.ComputeKNNOutlierScores<O,D>

All Implemented Interfaces:

InspectionUtilFrequentlyScanned, Parameterizable

@Reference(authors="E. Schubert, R. Wojdanowski, A. Zimek, H.-P. Kriegel",
           title="On Evaluation of Outlier Rankings and Outlier Scores",
           booktitle="Proc. 12th SIAM International Conference on Data Mining (SDM), Anaheim, CA, 2012.")
public class ComputeKNNOutlierScores<O,D extends NumberDistance<D,?>>
extends AbstractApplication

Application that runs a series of kNN-based algorithms on a data set, for building an ensemble in a second step. The output file consists of a label and one score value for each object. Reference:

E. Schubert, R. Wojdanowski, A. Zimek, H.-P. Kriegel
On Evaluation of Outlier Rankings and Outlier Scores
In Proceedings of the 12th SIAM International Conference on Data Mining (SDM), Anaheim, CA, 2012.

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
private static interface	ComputeKNNOutlierScores.AlgRunner Run an algorithm for a given k.
static class	ComputeKNNOutlierScores.Parameterizer<O,D extends NumberDistance<D,?>> Parameterization class.

Field Summary

Fields

Modifier and Type	Field and Description
(package private) ByLabelOutlier	bylabel By label outlier detection - reference
(package private) DistanceFunction<? super O,D>	distf Distance function to use
(package private) InputStep	inputstep Input step
private static Logging	logger Our logger class.
(package private) int	maxk Maximum value of k
(package private) File	outfile Output file
(package private) boolean	runabod Include ABOD in the experiments.
(package private) int	startk Starting value of k.
(package private) int	stepk k step size

Fields inherited from class de.lmu.ifi.dbs.elki.application.AbstractApplication

INFORMATION, INPUT_ID, OUTPUT_ID, STATIC_LOGGER

Constructor Summary

Constructors

Constructor and Description
ComputeKNNOutlierScores(boolean verbose, InputStep inputstep, DistanceFunction<? super O,D> distf, int startk, int stepk, int maxk, ByLabelOutlier bylabel, File outfile) Constructor.

Method Summary

Methods

Modifier and Type	Method and Description
(package private) void	detachResult(Database database, OutlierResult discardresult) Avoid that (future changes?)
static void	main(String[] args) Main method.
void	run() Runs the application.
private void	runForEachK(ComputeKNNOutlierScores.AlgRunner runner) Iterate over the k range.
(package private) void	writeResult(PrintStream out, DBIDs ids, OutlierResult result, ScalingFunction scaling, String label) Write a single output line.

Methods inherited from class de.lmu.ifi.dbs.elki.application.AbstractApplication

isVerbose, printErrorMessage, runCLIApplication, usage

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

Our logger class.

inputstep

final InputStep inputstep

Input step

distf

final DistanceFunction<? super O,D extends NumberDistance<D,?>> distf

Distance function to use

startk

final int startk

Starting value of k.

stepk

final int stepk

k step size

maxk

final int maxk

Maximum value of k

outfile

File outfile

Output file

bylabel

ByLabelOutlier bylabel

By label outlier detection - reference

runabod

boolean runabod

Include ABOD in the experiments.

Constructor Detail

ComputeKNNOutlierScores

public ComputeKNNOutlierScores(boolean verbose,
                       InputStep inputstep,
                       DistanceFunction<? super O,D> distf,
                       int startk,
                       int stepk,
                       int maxk,
                       ByLabelOutlier bylabel,
                       File outfile)

Constructor.

Parameters:

verbose - Verbose flag

inputstep - Input step

distf - Distance function

startk - Starting value of k

stepk - K step size

maxk - Maximum k value

bylabel - By label outlier (reference)

outfile - Output file

Method Detail

run

public void run()

Description copied from class: AbstractApplication

Runs the application.

Specified by:

run in class AbstractApplication

detachResult

void detachResult(Database database,
                OutlierResult discardresult)

Avoid that (future changes?) keep a reference to the result.

Parameters:

database - Database

discardresult - Result to discard.

writeResult

void writeResult(PrintStream out,
               DBIDs ids,
               OutlierResult result,
               ScalingFunction scaling,
               String label)

Write a single output line.

Parameters:

out - Output stream

ids - DBIDs

result - Outlier result

scaling - Scaling function.

label - Identification label

runForEachK

private void runForEachK(ComputeKNNOutlierScores.AlgRunner runner)

Iterate over the k range.

Parameters:

runner - Runner to run

main

public static void main(String[] args)

Main method.

Parameters:

args - Command line parameters.