HCLAS_KNN_CLASS: Latent class based modeling of binary outcomes

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

Latent class based modeling of binary outcomes

Description

Modeling a binary outcome via the the discovery of latent clusters. Each discovered latent cluster is modeled by the user provided fit function. Discovered clusters will be modeled by KNN or SVM.

Usage

	HLCM(formula = formula, 
	                data=NULL,
	                method=BSWiMS.model,
	                hysteresis = 0.1,
					classMethod=KNN_method,
					classModel.Control=NULL,
					minsize=10,
	                ...
					)

Arguments

formula	the base formula to extract the outcome
data	the data to be used for training the method
method	the binary classification function
hysteresis	the hysteresis shift for detecting wrongly classified subjects
classMethod	the function name for modeling the discovered latent clusters
classModel.Control	the parameters to be passed to the latent-class fitting function
minsize	the minimum size of the discovered clusters
...	parameters for the classification function

Value

original	The original model trained with all the dataset
alternativeModel	The model used to classify the wrongly classified samples
classModel	The method that models the latent class
accuracy	The original accuracy
selectedfeatures	The character vector of selected features
hysteresis	The used hysteresis
classSet	The discovered class label of each sample

Author(s)

Jose G. Tamez-Pena

See Also

class::knn

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