de.lmu.ifi.dbs.elki.math.linearalgebra.pca

Class AutotuningPCA

java.lang.Object

de.lmu.ifi.dbs.elki.math.linearalgebra.pca.PCARunner

de.lmu.ifi.dbs.elki.math.linearalgebra.pca.AutotuningPCA

@Reference(authors="Hans-Peter Kriegel, Peer Kr\u00f6ger, Erich Schubert, Arthur Zimek",
           title="A General Framework for Increasing the Robustness of PCA-based Correlation Clustering Algorithms",
           booktitle="Proc. 20th Intl. Conf. on Scientific and Statistical Database Management (SSDBM)",
           url="https://doi.org/10.1007/978-3-540-69497-7_27",
           bibkey="DBLP:conf/ssdbm/KriegelKSZ08")
public class AutotuningPCA
extends PCARunner

Performs a self-tuning local PCA based on the covariance matrices of given objects. At most the closest 'k' points are used in the calculation and a weight function is applied.

The number of points used depends on when the strong eigenvectors exhibit the clearest correlation.

Reference:

A General Framework for Increasing the Robustness of PCA-Based Correlation Clustering Algorithms
Hans-Peter Kriegel and Peer Kröger and Erich Schubert and Arthur Zimek
Proc. 20th Int. Conf. on Scientific and Statistical Database Management (SSDBM)

Since:

0.5.0

Author:

Erich Schubert

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
private static class	AutotuningPCA.Cand Candidate
static class	AutotuningPCA.Parameterizer Parameterization class.

Field Summary

Fields

Modifier and Type	Field and Description
private EigenPairFilter	filter Filter to select eigenvectors.

Fields inherited from class de.lmu.ifi.dbs.elki.math.linearalgebra.pca.PCARunner

covarianceMatrixBuilder

Constructor Summary

Constructors

Constructor and Description
AutotuningPCA(CovarianceMatrixBuilder covarianceMatrixBuilder, EigenPairFilter filter) Constructor.

Method Summary

Modifier and Type	Method and Description
private void	assertSortedByDistance(DoubleDBIDList results) Ensure that the results are sorted by distance.
private double	computeExplainedVariance(double[] eigenValues, int filteredEigenPairs) Compute the explained variance for a filtered EigenPairs.
PCAResult	processIds(DBIDs ids, Relation<? extends NumberVector> database) Run PCA on a collection of database IDs.
PCAResult	processQueryResult(DoubleDBIDList results, Relation<? extends NumberVector> database) Run PCA on a QueryResult Collection.
private static double[]	reversed(double[] a) Sort an array of doubles in descending order.

Methods inherited from class de.lmu.ifi.dbs.elki.math.linearalgebra.pca.PCARunner

getCovarianceMatrixBuilder, processCovarMatrix, processEVD, setCovarianceMatrixBuilder

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Detail

filter

private EigenPairFilter filter

Filter to select eigenvectors.

Constructor Detail

AutotuningPCA

public AutotuningPCA(CovarianceMatrixBuilder covarianceMatrixBuilder,
                     EigenPairFilter filter)

Constructor.

Parameters:

covarianceMatrixBuilder - Covariance matrix builder

filter - Filter to select eigenvectors

Method Detail

processIds

public PCAResult processIds(DBIDs ids,
                            Relation<? extends NumberVector> database)

Description copied from class: PCARunner

Run PCA on a collection of database IDs.

Overrides:

processIds in class PCARunner

Parameters:

ids - a collection of ids

database - the database used

Returns:

PCA result

processQueryResult

public PCAResult processQueryResult(DoubleDBIDList results,
                                    Relation<? extends NumberVector> database)

Description copied from class: PCARunner

Run PCA on a QueryResult Collection.

Overrides:

processQueryResult in class PCARunner

Parameters:

results - a collection of QueryResults

database - the database used

Returns:

PCA result

reversed

private static double[] reversed(double[] a)

Sort an array of doubles in descending order.

Parameters:

a - Values

Returns:

Values in descending order

computeExplainedVariance

private double computeExplainedVariance(double[] eigenValues,
                                        int filteredEigenPairs)

Compute the explained variance for a filtered EigenPairs.

Parameters:

eigenValues - Eigen values

filteredEigenPairs - Filtered eigenpairs

Returns:

explained variance by the strong eigenvectors.

assertSortedByDistance

private void assertSortedByDistance(DoubleDBIDList results)

Ensure that the results are sorted by distance.

Parameters:

results - Results to process