lti::competitiveAgglomeration Class Reference

This class implements the centroid clustering algorithm presented in "Clustering by Competitive Agglomeration" from Frigui and Krishnapuram, Pattern Recognition, Vol. More...

#include <ltiCompetitiveAgglomeration.h>

Inheritance diagram for lti::competitiveAgglomeration:

[legend]

Collaboration diagram for lti::competitiveAgglomeration:

[legend]

List of all members.

Classes
class	parameters
	The parameters for the class competitiveAgglomeration. More...
Public Member Functions
	competitiveAgglomeration ()
	competitiveAgglomeration (const parameters &par)
	competitiveAgglomeration (const competitiveAgglomeration &other)
virtual	~competitiveAgglomeration ()
virtual const char *	getTypeName () const
competitiveAgglomeration &	copy (const competitiveAgglomeration &other)
competitiveAgglomeration &	operator= (const competitiveAgglomeration &other)
virtual classifier *	clone () const
const parameters &	getParameters () const
bool	train (const lti::dmatrix &input)
bool	train (const lti::dmatrix &input, ivector &ids)

---

Detailed Description

This class implements the centroid clustering algorithm presented in "Clustering by Competitive Agglomeration" from Frigui and Krishnapuram, Pattern Recognition, Vol.

30, No. 7, pp. 1109-1119, 1997. The algorithm is a fuzzy clustering algorithm, which reduces a given partition to an optimal number of clusters. Here, the initial partition is generated by first performing a fuzzy-C-Means run on the data. Note: since the number of clusters will only be decreased by this algorithm, the fuzzy-C-Means parameters should be modified such that the number of clusters is chosen to be much larger than the expected number of clusters.

---

Constructor & Destructor Documentation

lti::competitiveAgglomeration::competitiveAgglomeration

(

)

Default constructor.

lti::competitiveAgglomeration::competitiveAgglomeration

(

const parameters &

par

)

Construct a classifier using the given parameters.

lti::competitiveAgglomeration::competitiveAgglomeration

(

const competitiveAgglomeration &

other

)

Copy constructor.

Parameters:

other

the object to be copied

virtual lti::competitiveAgglomeration::~competitiveAgglomeration

(

)

[virtual]

Destructor.

---

Member Function Documentation

virtual classifier* lti::competitiveAgglomeration::clone

(

)

const [virtual]

Returns a pointer to a clone of this classifier.

Implements lti::classifier.

competitiveAgglomeration& lti::competitiveAgglomeration::copy

(

const competitiveAgglomeration &

other

)

Copy data of "other" classifier.

Parameters:

other

the classifier to be copied

Returns:

a reference to this classifier object

Reimplemented from lti::centroidClustering.

const parameters& lti::competitiveAgglomeration::getParameters

(

)

const

Returns used parameters.

Reimplemented from lti::centroidClustering.

virtual const char* lti::competitiveAgglomeration::getTypeName

(

)

const [virtual]

Returns the name of this type ("competitiveAgglomeration").

Reimplemented from lti::centroidClustering.

competitiveAgglomeration& lti::competitiveAgglomeration::operator=

(

const competitiveAgglomeration &

other

)

Alias for copy member.

Parameters:

other

the classifier to be copied

Returns:

a reference to this classifier object

Reimplemented from lti::classifier.

bool lti::competitiveAgglomeration::train	(	const lti::dmatrix &	input,
		ivector &	ids
	)			[inline, virtual]

calls centroidClustering::train

Reimplemented from lti::centroidClustering.

References train().

bool lti::competitiveAgglomeration::train

(

const lti::dmatrix &

input

)

[virtual]

train clusters from given data.

Implements lti::centroidClustering.

Referenced by train().

---

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

• ltiCompetitiveAgglomeration.h