Use multiple feature selection algorithms to derive robust feature sets for two class classification problems.


[Up] [Top]

Documentation for package ‘bootfs’ version 1.0.6

Help Pages

bootFS-package Use multiple feature selection algorithms to derive robust feature sets for two class classification problems.
bootFS Use multiple feature selection algorithms to derive robust feature sets for two class classification problems.
bsPAMR Perform PAMR bootstrapping.
bsRFBORUTA Perform RFBORUTA bootstrapping.
bsSCAD Perform SCAD SVM bootstrapping.
cvPAMR Main wrapper to call PAMR crossvalidation.
cvRFBORUTA Crossvalidation for Random Forests with Boruta feature selection.
cvSCAD Crossvalidation for SCAD SVM classification and feature selection.
doBS Perform bootstrapped feature selection with multiple algorithms.
doCV Performance evaluation by crossvalidation for multiple classification algorithms.
drawheat Wrapper for heatmap drawing.
extractsignatures Helper for extracting all feature signatures from a bootstrapping result (single method).
importance_igraph Graphically represent the (co-)occurrences of a set of features, derived in a bootstrapped feature selection.
makeIG Create an importance graph from a bootstrapping result of a single classification method.
resultBS Summarise the results of a bootstrapping analysis.
simDataSet simDataSet - simulation of exemplary dataset