de.lmu.ifi.dbs.elki.algorithm

Class DependencyDerivator<V extends NumberVector<?>,D extends Distance<D>>

java.lang.Object

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

de.lmu.ifi.dbs.elki.algorithm.AbstractPrimitiveDistanceBasedAlgorithm<V,D,CorrelationAnalysisSolution<V>>

de.lmu.ifi.dbs.elki.algorithm.DependencyDerivator<V,D>

Type Parameters:

V - the type of FeatureVector handled by this Algorithm

D - the type of Distance used by this Algorithm

All Implemented Interfaces:

Algorithm, Parameterizable

@Title(value="Dependency Derivator: Deriving numerical inter-dependencies on data")
@Description(value="Derives an equality-system describing dependencies between attributes in a correlation-cluster")
@Reference(authors="E. Achtert, C. B\u00f6hm, H.-P. Kriegel, P. Kr\u00f6ger, A. Zimek",
           title="Deriving Quantitative Dependencies for Correlation Clusters",
           booktitle="Proc. 12th Int. Conf. on Knowledge Discovery and Data Mining (KDD \'06), Philadelphia, PA 2006.",
           url="http://dx.doi.org/10.1145/1150402.1150408")
public class DependencyDerivator<V extends NumberVector<?>,D extends Distance<D>>
extends AbstractPrimitiveDistanceBasedAlgorithm<V,D,CorrelationAnalysisSolution<V>>

Dependency derivator computes quantitatively linear dependencies among attributes of a given dataset based on a linear correlation PCA.

Reference:
E. Achtert, C. Böhm, H.-P. Kriegel, P. Kröger, A. Zimek: Deriving Quantitative Dependencies for Correlation Clusters.
In Proc. 12th Int. Conf. on Knowledge Discovery and Data Mining (KDD '06), Philadelphia, PA 2006.

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	DependencyDerivator.Parameterizer<V extends NumberVector<?>,D extends Distance<D>> Parameterization class.

Field Summary

Fields

Modifier and Type	Field and Description
static OptionID	DEPENDENCY_DERIVATOR_RANDOM_SAMPLE Flag to use random sample (use knn query around centroid, if flag is not set).
private static Logging	LOG The logger for this class.
private NumberFormat	nf Number format for output of solution.
static OptionID	OUTPUT_ACCURACY_ID Parameter to specify the threshold for output accuracy fraction digits, must be an integer equal to or greater than 0.
private PCAFilteredRunner<V>	pca Holds the object performing the pca.
private boolean	randomsample Flag for random sampling vs. kNN
static OptionID	SAMPLE_SIZE_ID Optional parameter to specify the threshold for the size of the random sample to use, must be an integer greater than 0.
private int	sampleSize Holds the value of SAMPLE_SIZE_ID.

Constructor Summary

Constructors

Constructor and Description
DependencyDerivator(PrimitiveDistanceFunction<V,D> distanceFunction, NumberFormat nf, PCAFilteredRunner<V> pca, int sampleSize, boolean randomsample) Constructor.

Method Summary

Methods

Modifier and Type	Method and Description
CorrelationAnalysisSolution<V>	generateModel(Relation<V> db, DBIDs ids) Runs the pca on the given set of IDs.
CorrelationAnalysisSolution<V>	generateModel(Relation<V> db, DBIDs ids, Vector centroid) Runs the pca on the given set of IDs and for the given centroid.
TypeInformation[]	getInputTypeRestriction() Get the input type restriction used for negotiating the data query.
protected Logging	getLogger() Get the (STATIC) logger for this class.
CorrelationAnalysisSolution<V>	run(Database database, Relation<V> relation) Computes quantitatively linear dependencies among the attributes of the given database based on a linear correlation PCA.

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

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

Field Detail

LOG

private static final Logging LOG

The logger for this class.

DEPENDENCY_DERIVATOR_RANDOM_SAMPLE

public static final OptionID DEPENDENCY_DERIVATOR_RANDOM_SAMPLE

Flag to use random sample (use knn query around centroid, if flag is not set).

OUTPUT_ACCURACY_ID

public static final OptionID OUTPUT_ACCURACY_ID

Parameter to specify the threshold for output accuracy fraction digits, must be an integer equal to or greater than 0.

SAMPLE_SIZE_ID

public static final OptionID SAMPLE_SIZE_ID

Optional parameter to specify the threshold for the size of the random sample to use, must be an integer greater than 0.

Default value: the size of the complete dataset

sampleSize

private final int sampleSize

Holds the value of SAMPLE_SIZE_ID.

pca

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

Holds the object performing the pca.

nf

private final NumberFormat nf

Number format for output of solution.

randomsample

private final boolean randomsample

Flag for random sampling vs. kNN

Constructor Detail

DependencyDerivator

public DependencyDerivator(PrimitiveDistanceFunction<V,D> distanceFunction,
                   NumberFormat nf,
                   PCAFilteredRunner<V> pca,
                   int sampleSize,
                   boolean randomsample)

Constructor.

Parameters:

distanceFunction - distance function

nf - Number format

pca - PCA runner

sampleSize - sample size

randomsample - flag for random sampling

Method Detail

run

public CorrelationAnalysisSolution<V> run(Database database,
                                 Relation<V> relation)

Computes quantitatively linear dependencies among the attributes of the given database based on a linear correlation PCA.

Parameters:

database - the database to run this DependencyDerivator on

relation - the relation to use

Returns:

the CorrelationAnalysisSolution computed by this DependencyDerivator

generateModel

public CorrelationAnalysisSolution<V> generateModel(Relation<V> db,
                                           DBIDs ids)

Runs the pca on the given set of IDs. The centroid is computed from the given ids.

Parameters:

db - the database

ids - the set of ids

Returns:

a matrix of equations describing the dependencies

generateModel

public CorrelationAnalysisSolution<V> generateModel(Relation<V> db,
                                           DBIDs ids,
                                           Vector centroid)

Runs the pca on the given set of IDs and for the given centroid.

Parameters:

db - the database

ids - the set of ids

centroid - the centroid

Returns:

a matrix of equations describing the dependencies

getInputTypeRestriction

public TypeInformation[] getInputTypeRestriction()

Description copied from class: AbstractAlgorithm

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

Specified by:

getInputTypeRestriction in interface Algorithm

Specified by:

getInputTypeRestriction in class AbstractAlgorithm<CorrelationAnalysisSolution<V extends NumberVector<?>>>

Returns:

Type restriction

getLogger

protected Logging getLogger()

Description copied from class: AbstractAlgorithm

Get the (STATIC) logger for this class.

Specified by:

getLogger in class AbstractAlgorithm<CorrelationAnalysisSolution<V extends NumberVector<?>>>

Returns:

the static logger