hybridEnsemble (version 1.7.9)

plot.CVhybridEnsemble: Plot the performance of the cross-validated Hybrid Ensemble

Description

This function plots the averaged ROC curve per combination method or the median predictive performance (Area under the ROC, sensitivity or specificity curve depending on what was used in the CVhybridEnsemble function).

Usage

# S3 method for CVhybridEnsemble
plot(x, y = NULL, ROCcurve = FALSE, averaging = "threshold", ...)

Arguments

x

An object of class CVhybridEnsemble

y

Not used

ROCcurve

TRUE or FALSE. Should the ROC curve be plotted or the median predictive performances?

averaging

For the ROC curve: "threshold" averaging, "horizontal" averaging, or "vertical" averaging.

...

Not used

Author

Michel Ballings, Dauwe Vercamer, Matthias Bogaert, and Dirk Van den Poel, Maintainer:

Details

In the output: 'RBGA' (Genetic Algorithm), 'DEOPT' (Differential Evolution), 'GENSA' (Generalized Simulated Annealing), 'MALSCHAINS' (Memetic Algorithm), 'PSOPTIM' (Particle Swarm), 'SOMA' (Self Organizing Migrating Algorithm), 'TABU' (Tabue Search), 'LHNNLS' (Lawson-Hanson Non-negative least squares), 'GINNLS' (Goldfarb-Idnani Non-negative least squares), 'NNloglik' (Non-negative binomial likelihood), 'MEAN' (Simple Mean), 'SB' (Single Best), 'AUTHORITY' (Authority Based method). SB names denote the single best for all cross-validation runs: RF= Random Forest, SV= Bagged Support Vector Machines, KF= Kernel Factory, AB=AdaBoost, LR=Bagged Logistic Regression, NN=Bagged Neural Networks, RoF= Rotation Forest, KN= K-Nearest Neighbors.

References

Ballings, M., Vercamer, D., Bogaert, M., Van den Poel, D.

See Also

hybridEnsemble, predict.hybridEnsemble, importance.hybridEnsemble, CVhybridEnsemble, summary.CVhybridEnsemble

Examples

data(Credit)

if (FALSE) {
CVhE <- CVhybridEnsemble(x=Credit[1:200,names(Credit) != 'Response'],
                    y=Credit$Response[1:200],
                    verbose=TRUE,
                    RF.ntree=50,
                    KF.rp=1,
                    AB.iter=50,
                    NN.size=5,
                    NN.decay=0,
                    SV.gamma = 2^-15,
                    SV.cost = 2^-5,
                    SV.degree=2,
                    SV.kernel='radial')

plot(x=CVhE,ROCcurve= FALSE)
plot(x=CVhE,ROCcurve= TRUE)
}