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.
HLCM(formula = formula,
data=NULL,
method=BSWiMS.model,
hysteresis = 0.1,
classMethod=KNN_method,
classModel.Control=NULL,
minsize=10,
...
)the base formula to extract the outcome
the data to be used for training the method
the binary classification function
the hysteresis shift for detecting wrongly classified subjects
the function name for modeling the discovered latent clusters
the parameters to be passed to the latent-class fitting function
the minimum size of the discovered clusters
parameters for the classification function
The original model trained with all the dataset
The model used to classify the wrongly classified samples
The method that models the latent class
The original accuracy
The character vector of selected features
The used hysteresis
The discovered class label of each sample
class::knn