Package elki.clustering.correlation

Class ERiC

java.lang.Object

elki.clustering.correlation.ERiC

All Implemented Interfaces:

Algorithm, ClusteringAlgorithm<Clustering<CorrelationModel>>

@Title("ERiC: Exploring Relationships among Correlation Clusters")
@Description("Performs the DBSCAN algorithm on the data using a special distance function taking into account correlations among attributes and builds a hierarchy that allows multiple inheritance from the correlation clustering result.")
@Reference(authors="Elke Achtert, Christian B\u00f6hm, Hans-Peter Kriegel, Peer Kr\u00f6ger, Arthur Zimek",
           title="On Exploring Complex Relationships of Correlation Clusters",
           booktitle="Proc. 19th Int. Conf. Scientific and Statistical Database Management (SSDBM 2007)",
           url="https://doi.org/10.1109/SSDBM.2007.21",
           bibkey="DBLP:conf/ssdbm/AchtertBKKZ07")
public class ERiC
extends java.lang.Object
implements ClusteringAlgorithm<Clustering<CorrelationModel>>

Performs correlation clustering on the data partitioned according to local correlation dimensionality and builds a hierarchy of correlation clusters that allows multiple inheritance from the clustering result.

Reference:

Elke Achtert, Christian Böhm, Hans-Peter Kriegel, Peer Kröger, Arthur Zimek
On Exploring Complex Relationships of Correlation Clusters
Proc. 19th Int. Conf. Scientific and Statistical Database Management (SSDBM 2007)

Since:

0.1

Author:

Elke Achtert

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
static class	ERiC.Par	Parameterization class.
static class	ERiC.Settings	Class to wrap the ERiC settings.

Nested classes/interfaces inherited from interface elki.Algorithm

Algorithm.Utils

Field Summary

Fields

Modifier and Type	Field	Description
private static Logging	LOG	The logger for this class.
private ERiC.Settings	settings	ERiC Settings.

Constructor Summary

Constructors

Constructor	Description
ERiC(ERiC.Settings settings)	Constructor.

Method Summary

Modifier and Type	Method	Description
private void	buildHierarchy(Clustering<CorrelationModel> clustering, java.util.List<java.util.List<Cluster<CorrelationModel>>> clusterMap, ERiCNeighborPredicate.Instance npred)
private java.util.List<java.util.List<Cluster<CorrelationModel>>>	extractCorrelationClusters(Clustering<Model> dbscanResult, Relation<? extends NumberVector> relation, int dimensionality, ERiCNeighborPredicate.Instance npred)	Extracts the correlation clusters and noise from the copac result and returns a mapping of correlation dimension to maps of clusters within this correlation dimension.
TypeInformation[]	getInputTypeRestriction()	Get the input type restriction used for negotiating the data query.
private boolean	isParent(ERiCNeighborPredicate.Instance npred, Cluster<CorrelationModel> parent, It<Cluster<CorrelationModel>> iter)	Returns true, if the specified parent cluster is a parent of one child of the children clusters.
Clustering<CorrelationModel>	run(Database database, Relation<? extends NumberVector> relation)	Performs the ERiC algorithm on the given database.

Methods inherited from class java.lang.Object

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

Methods inherited from interface elki.clustering.ClusteringAlgorithm

autorun

Field Detail

LOG

private static final Logging LOG

The logger for this class.

settings

private ERiC.Settings settings

ERiC Settings.

Constructor Detail

ERiC

public ERiC(ERiC.Settings settings)

Constructor.

Parameters:

settings - ERiC clustering settings

Method Detail

getInputTypeRestriction

public TypeInformation[] getInputTypeRestriction()

Description copied from interface: Algorithm

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

Specified by:

getInputTypeRestriction in interface Algorithm

Returns:

Type restriction

run

public Clustering<CorrelationModel> run(Database database,
                                        Relation<? extends NumberVector> relation)

Performs the ERiC algorithm on the given database.

Parameters:

relation - Relation to process

Returns:

Clustering result

extractCorrelationClusters

private java.util.List<java.util.List<Cluster<CorrelationModel>>> extractCorrelationClusters(Clustering<Model> dbscanResult,
                                                                                             Relation<? extends NumberVector> relation,
                                                                                             int dimensionality,
                                                                                             ERiCNeighborPredicate.Instance npred)

Extracts the correlation clusters and noise from the copac result and returns a mapping of correlation dimension to maps of clusters within this correlation dimension. Each cluster is defined by the basis vectors defining the subspace in which the cluster appears.

Parameters:

dbscanResult - DBSCAN clustering to use

relation - the database containing the objects

dimensionality - the dimensionality of the feature space

npred - ERiC predicate

Returns:

a list of clusters for each dimensionality

buildHierarchy

private void buildHierarchy(Clustering<CorrelationModel> clustering,
                            java.util.List<java.util.List<Cluster<CorrelationModel>>> clusterMap,
                            ERiCNeighborPredicate.Instance npred)

isParent

private boolean isParent(ERiCNeighborPredicate.Instance npred,
                         Cluster<CorrelationModel> parent,
                         It<Cluster<CorrelationModel>> iter)

Returns true, if the specified parent cluster is a parent of one child of the children clusters.

Parameters:

npred - Neighborhood predicate

parent - the parent to be tested

iter - the list of children to be tested

Returns:

true, if the specified parent cluster is a parent of one child of the children clusters, false otherwise