CompositionalML (version 1.0)

alpha-Boruta: \(\alpha\)-Boruta variable selection

Description

\(\alpha\)-Boruta variable selection.

Usage

alfa.boruta(y, x,  a = seq(-1, 1, by = 0.1), runs = 100 )

Value

A list with the results of the Boruta variable selection algortihm for each value of \(\alpha\)

as returned by the function "Boruta" of the package Boruta. The important elements are these (all the items returned by the function are of course included):

finalDecision

A factor of three values: "Confirmed", "Rejected" or "Tentative"", for each variable, containing the final result of the variable selection.

ImpHistory

A data frame of importances of variables gathered in each importance source run. Beside the importances, it contains maximal, mean and minimal importance of shadow variables in each run. Rejected attributes get -Inf importance. Set to NULL if holdHistory was given FALSE.

Arguments

y

The response variable, it can either be a factor (for classification) or a numeric vector (for regression). Depending on the nature of the response variable, the function will proceed with the necessary task.

x

A matrix with the compositional data.

a

A vector with a grid of values of the power transformation, it has to be between -1 and 1. If zero values are present it has to be greater than 0. If a=0, the isometric log-ratio transformation is applied.

runs

Maximal number of importance source runs. You may increase it to avoid variables being characterised as "Tentative".

Author

Michail Tsagris.

R implementation and documentation: Michail Tsagris .

Details

For each value of \(\alpha\), the compositional data are transformed and then the Boruta variable selection algorithm is applied.

References

Kursa M. B. and Rudnicki W. R. (2010). Feature Selection with the Boruta Package. Journal of Statistical Software, 36(11).

Tsagris M.T., Preston S. and Wood A.T.A. (2011). A data-based power transformation for compositional data. In Proceedings of the 4th Compositional Data Analysis Workshop, Girona, Spain. https://arxiv.org/pdf/1106.1451.pdf

See Also

alfa.rf

Examples

x <- as.matrix(iris[, 1:4])
x <- x/ rowSums(x)
y <- iris[, 5]
mod <- alfa.boruta(y, x, a = c(0, 0.5))
mod