rotationForest (version 0.1.3)

rotationForest: Binary classification with Rotation Forest (Rodriguez en Kuncheva, 2006)

Description

rotationForest implements an ensemble method where each base classifier (tree) is fit on the principal components of the variables of random partitions of the feature set.

Usage

rotationForest(x, y, K = round(ncol(x)/3, 0), L = 10, verbose = FALSE,
  ...)

Arguments

x
A data frame of predictors (numeric, or integer). Categorical variables need to be transformed to indicator (dummy) variables. At minimum x requires two columns.
y
A factor containing the response vector. Only {0,1} is allowed.
K
The number of variable subsets. The default is the value K that results in three features per subset.
L
The number of base classifiers (trees using the rpart package). The default is 10.
verbose
Boolean. Should information about the subsets be printed?
...
Arguments to rpart.control. First run library(rpart).

Value

An object of class rotationForest, which is a list with the following elements:
models
A list of trees.
loadings
A list of loadings.
columnnames
Column names of x.

References

Rodriguez, J.J., Kuncheva, L.I., 2006. Rotation forest: A new classifier ensemble method. IEEE Trans. Pattern Anal. Mach. Intell. 28, 1619-1630. doi:10.1109/TPAMI.2006.211

See Also

predict.rotationForest

Examples

Run this code
data(iris)
y <- as.factor(ifelse(iris$Species[1:100]=="setosa",0,1))
x <- iris[1:100,-5]
rF <- rotationForest(x,y)
predict(object=rF,newdata=x)

Run the code above in your browser using DataCamp Workspace