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

java.lang.Object
  de.lmu.ifi.dbs.elki.logging.AbstractLoggable
      de.lmu.ifi.dbs.elki.algorithm.AbstractAlgorithm<O,R>
          de.lmu.ifi.dbs.elki.algorithm.DistanceBasedAlgorithm<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<V,CorrelationAnalysisSolution<V>>, 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<V,?>,D extends Distance<D>>
extends DistanceBasedAlgorithm<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.

Author:

Arthur Zimek

Field Summary

static OptionID	DEPENDENCY_DERIVATOR_RANDOM_SAMPLE OptionID for RANDOM_SAMPLE_FLAG
NumberFormat	NF Number format for output of solution.
static OptionID	OUTPUT_ACCURACY_ID OptionID for OUTPUT_ACCURACY_PARAM
private IntParameter	OUTPUT_ACCURACY_PARAM Parameter to specify the threshold for output accuracy fraction digits, must be an integer equal to or greater than 0.
private PCAFilteredRunner<V,DoubleDistance>	pca Holds the object performing the pca.
private Flag	RANDOM_SAMPLE_FLAG Flag to use random sample (use knn query around centroid, if flag is not set).
static OptionID	SAMPLE_SIZE_ID OptionID for SAMPLE_SIZE_PARAM
private IntParameter	SAMPLE_SIZE_PARAM Optional parameter to specify the treshold for the size of the random sample to use, must be an integer greater than 0.
private Integer	sampleSize Holds the value of SAMPLE_SIZE_PARAM.

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

DISTANCE_FUNCTION_ID, DISTANCE_FUNCTION_PARAM

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

debug, logger

Constructor Summary

DependencyDerivator(Parameterization config)
Constructor, adhering to Parameterizable

Method Summary

CorrelationAnalysisSolution<V>	generateModel(Database<V> db, Collection<Integer> ids) Runs the pca on the given set of IDs.
CorrelationAnalysisSolution<V>	generateModel(Database<V> db, Collection<Integer> ids, V centroidDV) Runs the pca on the given set of IDs and for the given centroid.
CorrelationAnalysisSolution<V>	runInTime(Database<V> db) 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.DistanceBasedAlgorithm

getDistanceFactory, getDistanceFunction

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

isTime, isVerbose, run, setTime, setVerbose

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

DEPENDENCY_DERIVATOR_RANDOM_SAMPLE

public static final OptionID DEPENDENCY_DERIVATOR_RANDOM_SAMPLE

OptionID for RANDOM_SAMPLE_FLAG

OUTPUT_ACCURACY_ID

public static final OptionID OUTPUT_ACCURACY_ID

OptionID for OUTPUT_ACCURACY_PARAM

OUTPUT_ACCURACY_PARAM

private final IntParameter OUTPUT_ACCURACY_PARAM

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

Default value: 4

Key: -derivator.accuracy

SAMPLE_SIZE_ID

public static final OptionID SAMPLE_SIZE_ID

OptionID for SAMPLE_SIZE_PARAM

SAMPLE_SIZE_PARAM

private final IntParameter SAMPLE_SIZE_PARAM

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

Default value: the size of the complete dataset

Key: -derivator.sampleSize

sampleSize

private Integer sampleSize

Holds the value of SAMPLE_SIZE_PARAM.

RANDOM_SAMPLE_FLAG

private final Flag RANDOM_SAMPLE_FLAG

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

Key: -derivator.randomSample

pca

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

Holds the object performing the pca.

NF

public final NumberFormat NF

Number format for output of solution.

Constructor Detail

DependencyDerivator

public DependencyDerivator(Parameterization config)

Constructor, adhering to Parameterizable

Parameters:

config - Parameterization

Method Detail

runInTime

public CorrelationAnalysisSolution<V> runInTime(Database<V> db)
                                                                   throws IllegalStateException

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

Specified by:

runInTime in class AbstractAlgorithm<V extends NumberVector<V,?>,CorrelationAnalysisSolution<V extends NumberVector<V,?>>>

Parameters:

db - the database to run this DependencyDerivator on

Returns:

the CorrelationAnalysisSolution computed by this DependencyDerivator

Throws:

IllegalStateException - if the algorithm has not been initialized properly (e.g. the setParameters(String[]) method has been failed to be called).

generateModel

public CorrelationAnalysisSolution<V> generateModel(Database<V> db,
                                                    Collection<Integer> 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(Database<V> db,
                                                    Collection<Integer> ids,
                                                    V centroidDV)

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

Parameters:

db - the database

ids - the set of ids

centroidDV - the centroid

Returns:

a matrix of equations describing the dependencies