de.lmu.ifi.dbs.elki.preprocessing
Class FourCPreprocessor<D extends Distance<D>,V extends NumberVector<V,?>>

java.lang.Object
  de.lmu.ifi.dbs.elki.logging.AbstractLoggable
      de.lmu.ifi.dbs.elki.preprocessing.ProjectedDBSCANPreprocessor<D,V>
          de.lmu.ifi.dbs.elki.preprocessing.FourCPreprocessor<D,V>

Type Parameters:

D - Distance type

V - Vector type

All Implemented Interfaces:

Preprocessor<V>, Parameterizable

@Title(value="4C Preprocessor")
@Description(value="Computes the local dimensionality and locally weighted matrix of objects of a certain database according to the 4C algorithm.\nThe PCA is based on epsilon range queries.")
public class FourCPreprocessor<D extends Distance<D>,V extends NumberVector<V,?>>
extends ProjectedDBSCANPreprocessor<D,V>
implements Parameterizable

Preprocessor for 4C local dimensionality and locally weighted matrix assignment to objects of a certain database.

Author:

Arthur Zimek

Field Summary

private boolean	absolute Indicates whether delta is an absolute or a relative value.
private Flag	ABSOLUTE_FLAG Flag for marking parameter delta as an absolute value.
static double	DEFAULT_DELTA The default value for delta.
private double	delta Threshold for strong eigenpairs, can be absolute or relative.
private DoubleParameter	DELTA_PARAM Option string for parameter delta.
private PCAFilteredRunner<V,?>	pca The Filtered PCA Runner

Fields inherited from class de.lmu.ifi.dbs.elki.preprocessing.ProjectedDBSCANPreprocessor

DEFAULT_DISTANCE_FUNCTION, EPSILON_PARAM, rangeQueryDistanceFunction

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

debug, logger

Constructor Summary

FourCPreprocessor(Parameterization config)
Constructor, adhering to Parameterizable

Method Summary

protected void

runVarianceAnalysis(Integer id, List<DistanceResultPair<D>> neighbors, Database<V> database) This method implements the type of variance analysis to be computed for a given point.

Methods inherited from class de.lmu.ifi.dbs.elki.preprocessing.ProjectedDBSCANPreprocessor

run

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

ABSOLUTE_FLAG

private final Flag ABSOLUTE_FLAG

Flag for marking parameter delta as an absolute value.

DELTA_PARAM

private final DoubleParameter DELTA_PARAM

Option string for parameter delta.

DEFAULT_DELTA

public static final double DEFAULT_DELTA

The default value for delta.

See Also:

Constant Field Values

delta

private double delta

Threshold for strong eigenpairs, can be absolute or relative.

absolute

private boolean absolute

Indicates whether delta is an absolute or a relative value.

pca

private PCAFilteredRunner<V extends NumberVector<V,?>,?> pca

The Filtered PCA Runner

Constructor Detail

FourCPreprocessor

public FourCPreprocessor(Parameterization config)

Constructor, adhering to Parameterizable

Parameters:

config - Parameterization

Method Detail

runVarianceAnalysis

protected void runVarianceAnalysis(Integer id,
                                   List<DistanceResultPair<D>> neighbors,
                                   Database<V> database)

This method implements the type of variance analysis to be computed for a given point.

Example1: for 4C, this method should implement a PCA for the given point. Example2: for PreDeCon, this method should implement a simple axis-parallel variance analysis.

Specified by:

runVarianceAnalysis in class ProjectedDBSCANPreprocessor<D extends Distance<D>,V extends NumberVector<V,?>>

Parameters:

id - the given point

neighbors - the neighbors as query results of the given point

database - the database for which the preprocessing is performed