ClusteringTypes.h File Reference

Predefined types for clustering algorithm. More...

#include <IPSDKUtil/BaseCollections.h>
#include <IPSDKUtil/Tools/EnumMacros.h>
#include <IPSDKMath/LinearAlgebra/LinearAlgebraTypes.h>
#include <boost/shared_ptr.hpp>

Go to the source code of this file.

Classes
class	ipsdk::math::clustering::KernelKMeansClustering< KernelType >
class	ipsdk::math::clustering::SpectralClustering< KernelType >

Namespaces
	ipsdk
	Main namespace for IPSDK library.
	ipsdk::math
	Namespace agregating IPSDK mathematical routines and classes.

Typedefs
typedef std::vector< Vector >	ipsdk::math::clustering::VectorColl
	collection of vectors
typedef boost::shared_ptr< KMeansClustering >	ipsdk::math::clustering::KMeansClusteringPtr
	shared pointer to KMeansClustering class
typedef boost::shared_ptr< const KMeansClustering >	ipsdk::math::clustering::KMeansClusteringConstPtr
	shared pointer to const KMeansClustering class

Enumerations
enum	ipsdk::math::clustering::eValidityIndexType {   ipsdk::math::clustering::eVIT_Dunn, ipsdk::math::clustering::eVIT_DaviesBouldin, ipsdk::math::clustering::eVIT_RMSStd, ipsdk::math::clustering::eVIT_RS,   ipsdk::math::clustering::eVIT_SD }
	Enumerate describing a clustering validity indice type. More...

Detailed Description

Predefined types for clustering algorithm.

Author

E.Noirfalise

Date

2018/07/24

Copyright

2018 Reactiv'IP, all right reserved.

Enumeration Type Documentation

eValidityIndexType

enum ipsdk::math::clustering::eValidityIndexType

Enumerate describing a clustering validity indice type.

See also

Cluster Validity Measurement Techniques, Ferenc Kovács, Csaba Legány, Attila Babos, AIKED'06 Proceedings of the 5th WSEAS International Conference on Artificial Intelligence, Knowledge Engineering and Data Bases Pages 388-393

Note

This object is associated to a strong enumerate definition

See also

IPSDK_Enum

Enumerator
eVIT_Dunn	Dunn validity index. This index is a measure of ratio between minimum distance of two points belonging to different clusters and maximum distance of two points belonging to same cluster If a data set contains well-separated clusters, the distances among the clusters are usually large and the diameters of the clusters are expected to be small. Therefore larger value means better cluster configuration. See also https://en.wikipedia.org/wiki/Dunn_index
eVIT_DaviesBouldin	Davies Bouldin validity index. This index measures the average of similarity between each cluster and its most similar one As the clusters have to be compact and separated the lower Davies Bouldin index means better cluster configuration. See also https://en.wikipedia.org/wiki/Davies%E2%80%93Bouldin_index
eVIT_RMSStd	Root Mean Square Standard Deviation. This index measures the homogeneity of the clusters As the aim of the clustering process to identify homogenous groups the lower RMSSTD value means better clustering.
eVIT_RS	Root Squared. This index measures the dissimilarity of clusters (ie. the degree of homogeneity degree between groups) The values of RS range from 0 to 1 where 0 means there are no difference among the clusters and 1 indicates that there are significant difference among the clusters.
eVIT_SD	Scattering Distance. This index measures the the average scattering of clusters and total separation of clusters. Lower SD index means better cluster configuration as in this case the clusters are compacts and separated.