It implements Freund and Schapire's Adaboost.M1 algorithm and Breiman's Bagging algorithm using classification trees as individual classifiers. Once these classifiers have been trained, they can be used to predict on new data. Also, cross validation estimation of the error can be done. Since version 2.0 the function margins() is available to calculate the margins for these classifiers. Also a higher flexibility is achieved giving access to the rpart.control() argument of 'rpart'. Four important new features were introduced on version 3.0, AdaBoost-SAMME (Zhu et al., 2009) is implemented and a new function errorevol() shows the error of the ensembles as a function of the number of iterations. In addition, the ensembles can be pruned using the option 'newmfinal' in the predict.bagging() and predict.boosting() functions and the posterior probability of each class for observations can be obtained. Version 3.1 modifies the relative importance measure to take into account the gain of the Gini index given by a variable in each tree and the weights of these trees. Version 4.0 includes the margin-based ordered aggregation for Bagging pruning (Guo and Boukir, 2013) and a function to auto prune the 'rpart' tree. Moreover, three new plots are also available importanceplot(), plot.errorevol() and plot.margins(). Version 4.1 allows to predict on unlabeled data. Version 4.2 includes the parallel computation option for some of the functions.
|Author||Alfaro, Esteban; Gamez, Matias and Garcia, Noelia; with contributions from Li Guo|
|Maintainer||Esteban Alfaro <[email protected]>|
|License||GPL (>= 2)|
|Package repository||View on CRAN|
Install the latest version of this package by entering the following in R:
Any scripts or data that you put into this service are public.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.