de.lmu.ifi.dbs.elki.algorithm.clustering
Class ProjectedDBSCAN<V extends RealVector<V,?>>

java.lang.Object
  de.lmu.ifi.dbs.elki.logging.AbstractLoggable
      de.lmu.ifi.dbs.elki.utilities.optionhandling.AbstractParameterizable
          de.lmu.ifi.dbs.elki.algorithm.AbstractAlgorithm<V,Clustering<Model>>
              de.lmu.ifi.dbs.elki.algorithm.clustering.ProjectedDBSCAN<V>

Type Parameters:

V - the type of Realvector handled by this Algorithm

All Implemented Interfaces:

Algorithm<V,Clustering<Model>>, ClusteringAlgorithm<Clustering<Model>,V>, Parameterizable

Direct Known Subclasses:

FourC, PreDeCon

public abstract class ProjectedDBSCAN<V extends RealVector<V,?>>
extends AbstractAlgorithm<V,Clustering<Model>>
implements ClusteringAlgorithm<Clustering<Model>,V>

Provides an abstract algorithm requiring a VarianceAnalysisPreprocessor.

Author:

Arthur Zimek

Field Summary

static OptionID	DISTANCE_FUNCTION_ID OptionID for DISTANCE_FUNCTION_PARAM
protected ClassParameter<LocallyWeightedDistanceFunction<V,?>>	DISTANCE_FUNCTION_PARAM Parameter to specify the distance function to determine the distance between database objects, must extend AbstractLocallyWeightedDistanceFunction.
private LocallyWeightedDistanceFunction<V,?>	distanceFunction Holds the instance of the distance function specified by DISTANCE_FUNCTION_PARAM.
protected String	epsilon Holds the value of EPSILON_PARAM.
static OptionID	EPSILON_ID OptionID for EPSILON_PARAM
private PatternParameter	EPSILON_PARAM Parameter to specify the maximum radius of the neighborhood to be considered, must be suitable to LocallyWeightedDistanceFunction.
private int	lambda Holds the value of LAMBDA_PARAM.
static OptionID	LAMBDA_ID OptionID for LAMBDA_PARAM
private IntParameter	LAMBDA_PARAM Parameter to specify the intrinsic dimensionality of the clusters to find, must be an integer greater than 0.
protected int	minpts Holds the value of MINPTS_PARAM.
static OptionID	MINPTS_ID OptionID for MINPTS_PARAM
private IntParameter	MINPTS_PARAM Parameter to specify the threshold for minimum number of points in the epsilon-neighborhood of a point, must be an integer greater than 0.
private Set<Integer>	noise Holds a set of noise.
private Set<Integer>	processedIDs Holds a set of processed ids.
private Clustering<Model>	result Provides the result of the algorithm.
private List<List<Integer>>	resultList Holds a list of clusters found.

Fields inherited from class de.lmu.ifi.dbs.elki.utilities.optionhandling.AbstractParameterizable

optionHandler

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

debug, logger

Constructor Summary

protected

ProjectedDBSCAN() Provides the abstract algorithm for variance analysis based DBSCAN, adding parameters EPSILON_PARAM, MINPTS_PARAM, LAMBDA_PARAM, and DISTANCE_FUNCTION_PARAM to the option handler additionally to parameters of super class.

Method Summary

protected void	expandCluster(Database<V> database, Integer startObjectID, FiniteProgress objprog, IndefiniteProgress clusprog) ExpandCluster function of DBSCAN.
Clustering<Model>	getResult() Retrieve the result.
abstract Class<?>	preprocessorClass() Returns the class actually used as VarianceAnalysisPreprocessor.
protected Clustering<Model>	runInTime(Database<V> database) The run method encapsulated in measure of runtime.
List<String>	setParameters(List<String> args) Calls the super method and instantiates distanceFunction according to the value of parameter DISTANCE_FUNCTION_PARAM and sets additionally the values of the parameters EPSILON_PARAM MINPTS_PARAM, and LAMBDA_PARAM.

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.utilities.optionhandling.AbstractParameterizable

addOption, addParameterizable, addParameterizable, checkGlobalParameterConstraints, collectOptions, getAttributeSettings, getParameters, rememberParametersExcept, removeOption, removeParameterizable, shortDescription

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

Methods inherited from interface de.lmu.ifi.dbs.elki.algorithm.clustering.ClusteringAlgorithm

run

Methods inherited from interface de.lmu.ifi.dbs.elki.algorithm.Algorithm

getDescription, setTime, setVerbose

Methods inherited from interface de.lmu.ifi.dbs.elki.utilities.optionhandling.Parameterizable

checkGlobalParameterConstraints, collectOptions, getParameters, shortDescription

Field Detail

DISTANCE_FUNCTION_ID

public static final OptionID DISTANCE_FUNCTION_ID

OptionID for DISTANCE_FUNCTION_PARAM

DISTANCE_FUNCTION_PARAM

protected final ClassParameter<LocallyWeightedDistanceFunction<V extends RealVector<V,?>,?>> DISTANCE_FUNCTION_PARAM

Parameter to specify the distance function to determine the distance between database objects, must extend AbstractLocallyWeightedDistanceFunction.

Key: -projdbscan.distancefunction

Default value: EuclideanDistanceFunction

distanceFunction

private LocallyWeightedDistanceFunction<V extends RealVector<V,?>,?> distanceFunction

Holds the instance of the distance function specified by DISTANCE_FUNCTION_PARAM.

EPSILON_ID

public static final OptionID EPSILON_ID

OptionID for EPSILON_PARAM

EPSILON_PARAM

private final PatternParameter EPSILON_PARAM

Parameter to specify the maximum radius of the neighborhood to be considered, must be suitable to LocallyWeightedDistanceFunction.

Key: -projdbscan.epsilon

epsilon

protected String epsilon

Holds the value of EPSILON_PARAM.

LAMBDA_ID

public static final OptionID LAMBDA_ID

OptionID for LAMBDA_PARAM

LAMBDA_PARAM

private final IntParameter LAMBDA_PARAM

Parameter to specify the intrinsic dimensionality of the clusters to find, must be an integer greater than 0.

Key: -projdbscan.lambda

lambda

private int lambda

Holds the value of LAMBDA_PARAM.

MINPTS_ID

public static final OptionID MINPTS_ID

OptionID for MINPTS_PARAM

MINPTS_PARAM

private final IntParameter MINPTS_PARAM

Parameter to specify the threshold for minimum number of points in the epsilon-neighborhood of a point, must be an integer greater than 0.

Key: -projdbscan.minpts

minpts

protected int minpts

Holds the value of MINPTS_PARAM.

resultList

private List<List<Integer>> resultList

Holds a list of clusters found.

result

private Clustering<Model> result

Provides the result of the algorithm.

noise

private Set<Integer> noise

Holds a set of noise.

processedIDs

private Set<Integer> processedIDs

Holds a set of processed ids.

Constructor Detail

ProjectedDBSCAN

protected ProjectedDBSCAN()

Provides the abstract algorithm for variance analysis based DBSCAN, adding parameters EPSILON_PARAM, MINPTS_PARAM, LAMBDA_PARAM, and DISTANCE_FUNCTION_PARAM to the option handler additionally to parameters of super class.

Method Detail

runInTime

protected Clustering<Model> runInTime(Database<V> database)
                               throws IllegalStateException

Description copied from class: AbstractAlgorithm

The run method encapsulated in measure of runtime. An extending class needs not to take care of runtime itself.

Specified by:

runInTime in class AbstractAlgorithm<V extends RealVector<V,?>,Clustering<Model>>

Parameters:

database - the database to run the algorithm on

Returns:

the Result computed by this algorithm

Throws:

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

expandCluster

protected void expandCluster(Database<V> database,
                             Integer startObjectID,
                             FiniteProgress objprog,
                             IndefiniteProgress clusprog)

ExpandCluster function of DBSCAN.

Parameters:

database - the database to run the algorithm on

startObjectID - the object id of the database object to start the expansion with

objprog - the progress object for logging the current status

setParameters

public List<String> setParameters(List<String> args)
                           throws ParameterException

Calls the super method and instantiates distanceFunction according to the value of parameter DISTANCE_FUNCTION_PARAM and sets additionally the values of the parameters EPSILON_PARAM MINPTS_PARAM, and LAMBDA_PARAM.

The remaining parameters are passed to the distanceFunction.

Specified by:

setParameters in interface Parameterizable

Overrides:

setParameters in class AbstractAlgorithm<V extends RealVector<V,?>,Clustering<Model>>

Parameters:

args - parameters to set the attributes accordingly to

Returns:

a list containing the unused parameters

Throws:

ParameterException - in case of wrong parameter-setting

preprocessorClass

public abstract Class<?> preprocessorClass()

Returns the class actually used as VarianceAnalysisPreprocessor.

Returns:

the class actually used as VarianceAnalysisPreprocessor

getResult

public Clustering<Model> getResult()

Description copied from interface: ClusteringAlgorithm

Retrieve the result.

Specified by:

getResult in interface Algorithm<V extends RealVector<V,?>,Clustering<Model>>

Specified by:

getResult in interface ClusteringAlgorithm<Clustering<Model>,V extends RealVector<V,?>>

Returns:

the result of the algorithm