Build, Deploy and Evaluate Hybrid Ensembles

Functions to build and deploy a hybrid ensemble consisting of eight different sub-ensembles: bagged logistic regressions, random forest, stochastic boosting, kernel factory, bagged neural networks, bagged support vector machines, rotation forest, and bagged k-nearest neighbors. Functions to cross-validate the hybrid ensemble and plot and summarize the results are also provided. There is also a function to assess the importance of the predictors.


Planned changes in next versions:

-parameter to yield prediction vectors of sub-ensembles and also summary and plots -add parameter to use percentile ranks instead of calibration -incorporate statistical test in table -mechanism to compute and plot variance of predictions for all sub-ensembles and meta ensemble

Change log:

  • Version 1.0.0: May, 26 2015

-added a seventh base classifier: rotation forest -add an eight base classifier: bagged nearest neighbors -added parameters for size of sub-ensembles -added oversampling to alleviate problems related to class imbalance and subsequent subsetting -added a filter parameter to remove near constants that often produced problems in subsetting. -added error handling -added formal automated tests -refactored for faster and shorter code -changed to [-1,1] scaling in neural networks -made sure that no (near) constants can be produced in bagging procedures -corrected typesetting issues in documentation

  • Version 0.1.1: March 2014
  • Fixed small bug that occurred in rare cases
  • Version 0.1.0: December 2013
  • Package submitted to CRAN with main functions hybridEnsemble, predict, importance, CVhybridEnsemble, plot, summary

Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.


1.0.0 by Michel Ballings, 4 years ago

Browse source code at

Authors: Michel Ballings , Dauwe Vercamer , and Dirk Van den Poel

Documentation:   PDF Manual  

GPL (>= 2) license

Imports randomForest, kernelFactory, ada, rpart, ROCR, nnet, e1071, NMOF, GenSA, Rmalschains, pso, AUC, soma, genalg, reportr, nnls, quadprog, tabuSearch, rotationForest, FNN, glmnet

Suggests testthat

See at CRAN