Package elki.index.tree.betula

Class CFTree.Factory<L extends ClusterFeature>

java.lang.Object

elki.index.tree.betula.CFTree.Factory<L>

Enclosing class:

CFTree<L extends ClusterFeature>

public static class CFTree.Factory<L extends ClusterFeature>
extends java.lang.Object

CF-Tree Factory.

Author:

Erich Schubert

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
static class	CFTree.Factory.Par<L extends ClusterFeature>	Parameterization class for CFTrees.

Field Summary

Fields

Modifier and Type	Field	Description
(package private) CFDistance	abs	BIRCH distance function to use for point absorption
(package private) int	branchingFactor	Maximum branching factor of CFTree.
(package private) CFDistance	dist	BIRCH distance function to use
(package private) ClusterFeature.Factory<L>	factory	Cluster feature factory
(package private) double	maxleaves	Maximum number of leaves (absolute or relative)
(package private) CFTree.Threshold	tCriterium	Threshold heuristic strategy.
(package private) double	threshold	Cluster merge threshold.

Constructor Summary

Constructors

Constructor	Description
Factory(ClusterFeature.Factory<L> factory, CFDistance dist, CFDistance abs, double threshold, int branchingFactor, double maxleaves, CFTree.Threshold tCriterium)	Constructor.

Method Summary

Modifier and Type	Method	Description
CFTree<L>	newTree(DBIDs ids, Relation<? extends NumberVector> relation, boolean storeIds)	Make a new tree.

Methods inherited from class java.lang.Object

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

Field Detail

factory

ClusterFeature.Factory<L extends ClusterFeature> factory

Cluster feature factory

dist

CFDistance dist

BIRCH distance function to use

abs

CFDistance abs

BIRCH distance function to use for point absorption

threshold

double threshold

Cluster merge threshold.

branchingFactor

int branchingFactor

Maximum branching factor of CFTree.

maxleaves

double maxleaves

Maximum number of leaves (absolute or relative)

tCriterium

CFTree.Threshold tCriterium

Threshold heuristic strategy.

Constructor Detail

Factory

public Factory(ClusterFeature.Factory<L> factory,
               CFDistance dist,
               CFDistance abs,
               double threshold,
               int branchingFactor,
               double maxleaves,
               CFTree.Threshold tCriterium)

Constructor.

Parameters:

factory - Cluster feature factory

dist - Distance function to choose nearest

abs - Absorption criterion

threshold - Distance threshold

branchingFactor - Maximum branching factor

maxleaves - Maximum number of leaves

tCriterium - threshold adjustment rule

Method Detail

newTree

public CFTree<L> newTree(DBIDs ids,
                         Relation<? extends NumberVector> relation,
                         boolean storeIds)

Make a new tree.

Parameters:

ids - DBIDs to insert

relation - Data relation

Returns:

New tree