lti::competitiveAgglomeration::parameters Class Reference

The parameters for the class competitiveAgglomeration. More...

#include <ltiCompetitiveAgglomeration.h>

List of all members.

Public Member Functions
	parameters ()
	parameters (const parameters &other)
	~parameters ()
const char *	getTypeName () const
parameters &	copy (const parameters &other)
parameters &	operator= (const parameters &other)
virtual classifier::parameters *	clone () const
virtual bool	write (ioHandler &handler, const bool complete=true) const
virtual bool	read (ioHandler &handler, const bool complete=true)
Public Attributes
double	cardinalityFactor
double	timeConstant
int	initialNumberOfClusters
int	initialIterations
double	minimumCardinality
double	fuzzifier
int	maxIterations
double	convergenceThreshold

---

Detailed Description

The parameters for the class competitiveAgglomeration.

---

Constructor & Destructor Documentation

lti::competitiveAgglomeration::parameters::parameters

(

)

Default constructor.

lti::competitiveAgglomeration::parameters::parameters

(

const parameters &

other

)

Copy constructor.

Parameters:

other

the parameters object to be copied

lti::competitiveAgglomeration::parameters::~parameters

(

)

Destructor.

---

Member Function Documentation

virtual classifier::parameters* lti::competitiveAgglomeration::parameters::clone

(

)

const [virtual]

Returns a pointer to a clone of the parameters.

parameters& lti::competitiveAgglomeration::parameters::copy

(

const parameters &

other

)

Copy the contents of a parameters object.

Parameters:

other

the parameters object to be copied

Returns:

a reference to this parameters object

const char* lti::competitiveAgglomeration::parameters::getTypeName

(

)

const

Returns name of this type.

parameters& lti::competitiveAgglomeration::parameters::operator=

(

const parameters &

other

)

Copy the contents of a parameters object.

Parameters:

other

the parameters object to be copied

Returns:

a reference to this parameters object

virtual bool lti::competitiveAgglomeration::parameters::read	(	ioHandler &	handler,
		const bool	complete = true
	)			[virtual]

Read the parameters from the given ioHandler.

Parameters:

	handler	the ioHandler to be used
	complete	if true (the default) the enclosing begin/end will be also written, otherwise only the data block will be written.

Returns:

true if write was successful

virtual bool lti::competitiveAgglomeration::parameters::write	(	ioHandler &	handler,
		const bool	complete = true
	)			const [virtual]

Write the parameters in the given ioHandler.

Parameters:

	handler	the ioHandler to be used
	complete	if true (the default) the enclosing begin/end will be also written, otherwise only the data block will be written.

Returns:

true if write was successful

---

Member Data Documentation

double lti::competitiveAgglomeration::parameters::cardinalityFactor

The cardinality factor defines the importance of merging nearby clusters relative to covering the data optimally.

Larger values will result in fewer clusters, with a value of 0.0 no cluster will be discarded, thus leading to the same result as the fuzzy C-Means algorithm. Default value is 5.0

double lti::competitiveAgglomeration::parameters::convergenceThreshold

The algorithm converges, when the centroids remain stable.

If the sum of the L2 distance of the relative movements of the centroids in one iteration is below the convergence threshold, the algorithm terminates. Default: 0.02

double lti::competitiveAgglomeration::parameters::fuzzifier

Bias the algorithm either towards hard clustering (nearby 1) or fuzzy clustering (bigger 1); this parameter must be bigger than 1.

Default is 2.0

int lti::competitiveAgglomeration::parameters::initialIterations

The number of iterations to be used for fuzzy c-means pre-classification.

Default: 10.

int lti::competitiveAgglomeration::parameters::initialNumberOfClusters

The initial number of clusters must be larger than the expected optimal number of clusters.

Default is 20.

int lti::competitiveAgglomeration::parameters::maxIterations

The maximum number of iterations serves as one of two convergence criteria, the other one being convergenceThreshold.

Default value is 100.

double lti::competitiveAgglomeration::parameters::minimumCardinality

The cardinality describes the amount of data that is being covered by the cluster.

If the cardinality of a cluster drops below a given threshold, it will be discarded. The default value for this threshold is 5.0

double lti::competitiveAgglomeration::parameters::timeConstant

As the cardinality factor relies heavily on the data, the distance measure and the number of clusters, a decay function is used that starts with the given cardinality factor and exponentially decreases the value with each iteration: exp(-iteration / timeConstant).

This should result in finding the optimal number of clusters faster. The default time constant is 10.0

---

The documentation for this class was generated from the following file:

• ltiCompetitiveAgglomeration.h