bootfs: Derive Robust Feature Sets for Two or Multiclass Classification Problems

Share:

In order to select a set of features used for successful classification of two or more groups of samples, multiple classification and feature selection algorithms are utilised. By combining the results of all methods and applying a bootstrapping approach a robust set of features with high power to distinguish the sample groups is selected.

Author
Christian Bender
Date of publication
2016-02-07 14:37:37
Maintainer
Christian Bender <benderpkg@gmail.com>
License
GPL (>=2)
Version
1.4.5

View on R-Forge

Man pages

align_feat_boruta
Create matrix with aligned feature sets from Boruta results.
align_features
Create matrix with aligned feature sets from SVM SCAD...
bootfs-package
Use multiple feature selection algorithms to derive robust...
bootstrap_roc
Create roc curves from bootstrapping classification results.
bsGBM
Wrapper for GBM bootstrapping
bsPAMR
Perform PAMR bootstrapping.
bsRFBORUTA
Perform RFBORUTA bootstrapping.
bsSCAD
Perform SCAD SVM bootstrapping.
bstr_multi
SVM SCAD bootstrapping classification and feature selection.
control_params
Create control parameter object for the classifiers
cvGBM
Make a crossvalidation using GBM.
cv_gbmclass
Internal crossvalidation method for GBM classification
cv_pamclass
Wrapper for PAMR classification call.
cvPAMR
Main wrapper to call PAMR crossvalidation.
cv_penalizedSVM
SVM SCAD crossvalidation.
cvRFBORUTA
Crossvalidation for Random Forests with Boruta feature...
cv_rfclass
Wrapper for RF Boruta classification crossvalidation.
cvSCAD
Crossvalidation for SCAD SVM classification and feature...
dist.eucsq
Squared euclidean distance metric. Used for ward clustering.
doBS
Perform bootstrapped feature selection with multiple...
doCV
Performance evaluation by crossvalidation for multiple...
drawheat
Wrapper for heatmap drawing.
extract_feat_boruta
Feature extraction for Boruta bootstrapping results.
extract_feat_pam
Feature extraction for PAMR bootstrapping results.
extract_feature_rankings
Feature ranking extraction for SVM SCAD bootstrapping...
extract_features
Helper for feature extraction.
extract_features_rf_boruta
Helper for feature extraction for RF Boruta.
extractsignatures
Helper for extracting all feature signatures from a...
fitGBM
Fit a Gradient Boosting Machine model.
gbm_multi
Calling function to GBM bootstrapping
get_ellipsis_a
Helper for importance graph drawing.
get_feature_ranking
Get a feature ranking. Helper for SVM SCAD feature...
get_pam_features
Helper for PAMR feature extraction.
importance_igraph
Graphically represent the (co-)occurrences of a set of...
layout.ellipsis
Layout function creating an ellipsis.
makeIG
Create an importance graph from a bootstrapping result of a...
resultBS
Summarise the results of a bootstrapping analysis.
resultCV
create a result plot for all performed crossvalidations
rf_multi
Wrapper for RF Boruta bootstrapping.
roc
Draw ROC curve.
run_pam
Wrapper for PAMR bootstrapping classification.
select_bootstrap_data
Select bootstrapping samples preserving group member...
select_cv_balanced
Balanced training/test set selection for crossvalidation.
select_threshold
Find optimal threshold for PAMR feature selection.
simDataSet
simDataSet - simulation of exemplary dataset
svmclass
Wrapper for SVM SCAD classification call.
ward
Clustering function using ward clustering.

Files in this package

bootfs/DESCRIPTION
bootfs/NAMESPACE
bootfs/R
bootfs/R/align_feat_boruta.R
bootfs/R/align_features.R
bootfs/R/bootstrap_roc.R
bootfs/R/bsGBM.R
bootfs/R/bsPAMR.R
bootfs/R/bsRFBORUTA.R
bootfs/R/bsSCAD.R
bootfs/R/bs_methods.R
bootfs/R/control_params.R
bootfs/R/cvGBM.R
bootfs/R/cvPAMR.R
bootfs/R/cvRFBORUTA.R
bootfs/R/cvSCAD.R
bootfs/R/cv_gbmclass.R
bootfs/R/cv_pamclass.R
bootfs/R/cv_penalizedSVM.R
bootfs/R/cv_rfclass.R
bootfs/R/dist.eucsq.R
bootfs/R/doBS.R
bootfs/R/doCV.R
bootfs/R/drawheat.R
bootfs/R/extract_feat_boruta.R
bootfs/R/extract_feat_pam.R
bootfs/R/extract_feature_rankings.R
bootfs/R/extract_features.R
bootfs/R/extract_features_rf_boruta.R
bootfs/R/extractsignatures.R
bootfs/R/get_ellipsis_a.R
bootfs/R/get_feature_ranking.R
bootfs/R/get_pam_features.R
bootfs/R/importance_igraph.R
bootfs/R/layout.ellipsis.R
bootfs/R/makeIG.R
bootfs/R/resultBS.R
bootfs/R/resultCV.R
bootfs/R/roc.R
bootfs/R/select_bootstrap_data.R
bootfs/R/select_cv_balanced.R
bootfs/R/select_threshold.R
bootfs/R/simDataSet.R
bootfs/R/ward.R
bootfs/build
bootfs/build/vignette.rds
bootfs/inst
bootfs/inst/doc
bootfs/inst/doc/bootfs.R
bootfs/inst/doc/bootfs.Rnw
bootfs/inst/doc/bootfs.pdf
bootfs/man
bootfs/man/align_feat_boruta.Rd
bootfs/man/align_features.Rd
bootfs/man/bootfs-package.Rd
bootfs/man/bootstrap_roc.Rd
bootfs/man/bsGBM.Rd
bootfs/man/bsPAMR.Rd
bootfs/man/bsRFBORUTA.Rd
bootfs/man/bsSCAD.Rd
bootfs/man/bstr_multi.Rd
bootfs/man/control_params.Rd
bootfs/man/cvGBM.Rd
bootfs/man/cvPAMR.Rd
bootfs/man/cvRFBORUTA.Rd
bootfs/man/cvSCAD.Rd
bootfs/man/cv_gbmclass.Rd
bootfs/man/cv_pamclass.Rd
bootfs/man/cv_penalizedSVM.Rd
bootfs/man/cv_rfclass.Rd
bootfs/man/dist.eucsq.Rd
bootfs/man/doBS.Rd
bootfs/man/doCV.Rd
bootfs/man/drawheat.Rd
bootfs/man/extract_feat_boruta.Rd
bootfs/man/extract_feat_pam.Rd
bootfs/man/extract_feature_rankings.Rd
bootfs/man/extract_features.Rd
bootfs/man/extract_features_rf_boruta.Rd
bootfs/man/extractsignatures.Rd
bootfs/man/fitGBM.Rd
bootfs/man/gbm_multi.Rd
bootfs/man/get_ellipsis_a.Rd
bootfs/man/get_feature_ranking.Rd
bootfs/man/get_pam_features.Rd
bootfs/man/importance_igraph.Rd
bootfs/man/layout.ellipsis.Rd
bootfs/man/makeIG.Rd
bootfs/man/resultBS.Rd
bootfs/man/resultCV.Rd
bootfs/man/rf_multi.Rd
bootfs/man/roc.Rd
bootfs/man/run_pam.Rd
bootfs/man/select_bootstrap_data.Rd
bootfs/man/select_cv_balanced.Rd
bootfs/man/select_threshold.Rd
bootfs/man/simDataSet.Rd
bootfs/man/svmclass.Rd
bootfs/man/ward.Rd
bootfs/vignettes
bootfs/vignettes/bootfs.Rnw
bootfs/vignettes/impgraph1.pdf
bootfs/vignettes/impgraph2.pdf
bootfs/vignettes/references.bib
bootfs/vignettes/resultbsmulti.pdf
bootfs/vignettes/resultcvmulti.pdf
bootfs/vignettes/rocpam.pdf
bootfs/vignettes/rocrf.pdf
bootfs/vignettes/rocscad.pdf