GMVEBSWiMS: Hybrid Hierarchical Modeling with GMVE and BSWiMS

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

Description

This function returns the BSWiMS supervised-classifier present at each one of the GMVE unsupervised Gaussian data clusters

Usage

   GMVEBSWiMS(formula = formula,
            data=NULL,
            GMVE.control = list(p.threshold = 0.95,p.samplingthreshold = 0.5),
            ...
   )

Arguments

formula	An object of class formula with the formula to be fitted
data	A data frame where all variables are stored in different columns
GMVE.control	Control parameters of the GMVECluster function
...	Parameters to be passed to the BSWiMS.model function

Details

First, the function calls the BSWiMS function that returns the relevant features associated with the outcome. Then, it calls the GMVE clustering algorithm (GMVECluster) that returns a relevant data partition based on Gaussian clusters. Finally, the function will execute the BSWiMS.model classification function on each cluster returned by GMVECluster.

Value

features	The character vector with the releavant BSWiMS features.
cluster	The GMVECluster object
models	The list of BSWiMS.model models per cluster

Author(s)

Jose G. Tamez-Pena

Examples

   ## Not run: 
   # Get the Sonar data set
      library(mlbench)
      data(Sonar)
      Sonar$Class <- 1*(Sonar$Class == "M")
     #Train hierachical classifier
      mc <- GMVEBSWiMS(Class~.,Sonar)
     #report the classification
      pb <- predict(mc,Sonar)
      print(table(1*(pb>0.0),Sonar$Class))
   
## End(Not run)

FRESA.CAD documentation built on Jan. 13, 2021, 3:39 p.m.