GMVECluster: Set Clustering using the Generalized Minimum Volume Ellipsoid...

In FRESA.CAD: Feature Selection Algorithms for Computer Aided Diagnosis

Set Clustering using the Generalized Minimum Volume Ellipsoid (GMVE)

Description

The Function will return the set of Gaussian Ellipsoids that best model the data

Usage

	GMVECluster(dataset, 
	            p.threshold=0.975,
	            samples=10000,
	            p.samplingthreshold=0.50,
	            sampling.rate = 3,
	            jitter=TRUE,
	            tryouts=25,
	            pca=TRUE,
	            verbose=FALSE)

Arguments

dataset	The data set to be clustered
p.threshold	The p-value threshold of point acceptance into a set.
samples	If the set is large, The number of random samples
p.samplingthreshold	Defines the maximum distance between set candidate points
sampling.rate	Uniform sampling rate for candidate clusters
jitter	If true, will jitter the data set
tryouts	The number of cluster candidates that will be analyed per sampled point
pca	If TRUE, it will use the PCA transform for dimension reduction
verbose	If true it will print the clustering evolution

Details

Implementation of the GMVE clustering algorithm as proposed by Jolion et al. (1991).

Value

cluster	The numeric vector with the cluster label of each point
classification	The numeric vector with the cluster label of each point
centers	The list of cluster centers
covariances	The list of cluster covariance
robCov	The list of robust covariances per cluster
k	The number of discovered clusters
features	The characer vector with the names of the features used
jitteredData	The jittered dataset

Author(s)

Jose G. Tamez-Pena

References

Jolion, Jean-Michel, Peter Meer, and Samira Bataouche. "Robust clustering with applications in computer vision." IEEE Transactions on Pattern Analysis & Machine Intelligence 8 (1991): 791-802.

FRESA.CAD documentation built on Nov. 25, 2023, 1:07 a.m.