feature.selection: Selection of features

In pavhrablis/benchmarkFS: What the Package Does (One Line, Title Case)

Description

Selection of features

Usage

feature.selection(
  x,
  y,
  methods,
  connection.method = "sum",
  params = list(adjust = "SGoF", feature.number = 10, alpha = 0.05)
)

Arguments

x	input data where columns are variables and rows are observations (all numeric)
y	decision variable as a boolean vector of length equal to number of observations
connection.method	method of combining selection results if more than one method is selected, options 'sum' for the sum of all selected filtration methods or 'intersect' for part of the common
params	A list parameters required for this method of feature selection
method	the name of the selection method for traits available in this library

Details

Uses the methods of selection of features available in this library

Value

A new data.frame consisting of the selected functions and the decision variable

Examples

class <- data$class
data$class <- NULL

feature.selection(x = data,
                  y = class,
                  methods = c('fs.utest','fs.mdfs.1D', 'fs.mrmr', 'fs.mcfs'),
                  connection.method = 'sum',
                  params = list(feature.number=100, adjust = 'holm', alpha = 0.05))

pavhrablis/benchmarkFS documentation built on Feb. 5, 2021, 12:47 a.m.