jMEF.Clustering Class Reference

List of all members.

Public Types
enum	CLUSTERING_TYPE { LEFT_SIDED, RIGHT_SIDED, SYMMETRIC }
	Type of the Bregman divergence used. More...
Static Public Member Functions
static Parameter	getCenterOfMass (MixtureModel f)
	Computes the center of mass (right-sided centroid) of a mixture model f.
static Parameter	getGeneralizedCentroid (ExponentialFamily EF, MixtureModel f)
	Computes the generalized centroid (left-sided centroid) of a mixture model f.
static Parameter	getSymmetricCentroid (ExponentialFamily EF, MixtureModel f)
	Computes the symmetric centroid of a mixture model f approximated by the geodesic walk algorithm.
static Parameter	getCentroid (MixtureModel f, Clustering.CLUSTERING_TYPE type)
	Computes the centroid of a mixture model.

---

Detailed Description

Author:

Vincent Garcia

Frank Nielsen

Version:

1.0

License

See file LICENSE.txt

Description

This class provides the functions to compute the centroid of a mixture of exponential families. The centroid, depending on the non-symmetric Bregman divergence, can be:

• right-sided,
• left-sided,
• symmetric.

---

Member Enumeration Documentation

enum jMEF::Clustering::CLUSTERING_TYPE

Type of the Bregman divergence used.

Enumerator:

LEFT_SIDED
RIGHT_SIDED
SYMMETRIC

---

Member Function Documentation

static Parameter jMEF.Clustering.getCenterOfMass

(

MixtureModel

f

)

[static]

Computes the center of mass (right-sided centroid) of a mixture model f.

Parameters:

f

mixture model given in natural parameters

Returns:

center of mass of f

static Parameter jMEF.Clustering.getCentroid	(	MixtureModel	f,
		Clustering.CLUSTERING_TYPE	type
	)			[static]

Computes the centroid of a mixture model.

This centroid is sided (right- or left-sided) or is symmetric.

Parameters:

	f	mixture model
	type	type of Bregman divergence used (right-sided, left-sided, or symmetric)

Returns:

sided or symmetric centroid of f

static Parameter jMEF.Clustering.getGeneralizedCentroid	(	ExponentialFamily	EF,
		MixtureModel	f
	)			[static]

Computes the generalized centroid (left-sided centroid) of a mixture model f.

This centroid depends on the exponential family used in the mixture model.

Parameters:

	EF	exponential family
	f	mixture model given in natural parameters

Returns:

generalized centroid of f

static Parameter jMEF.Clustering.getSymmetricCentroid	(	ExponentialFamily	EF,
		MixtureModel	f
	)			[static]

Computes the symmetric centroid of a mixture model f approximated by the geodesic walk algorithm.

This centroid depends on the exponential family used in the mixture model.

Parameters:

	EF	exponential family
	f	mixture model given in natural parameters

Returns:

symmetric centroid of f

---

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

• Clustering.java