Feature selection by the Relief Algorithm for datasets containing nominal features

Description

This function applies the RELIEF Algorithm to datasets containing nominal attributes.

Usage

1
reliefcat(data, nosample, threshold, vnom, repet)

Arguments

data

The name of the dataset

nosample

The size of the sample drawn and used to update the relevance of each feature. Usually it is equal to the total number of instances.

threshold

The threshold for choosing the relevant features

vnom

A vector of indices indicating the nominal features

repet

The number of the repetitions. It is recommended to use at most 10 repetitions. If the nosample=number of instances then set repet=1

Author(s)

Edgar Acuna

See Also

relief

Want to suggest features or report bugs for rdrr.io? Use the GitHub issue tracker.