feature.boruta.selected: Provide all Non-Rejected Boruta Features

In MarcoNiemann/kaggle_house: Analysing House Prices in Ames, Iowa and Build a Sales Price Prediction Model

Description

This method combines the features that have either been confirmed or have been marked as tentative in the Boruta algorithm into one variable.

Usage

feature.boruta.selected(boruta)

Arguments

boruta

The Boruta object to which the Selected variable should be attached.

Details

This method usually is not meant to be applied by a package user. In case it is still used to obtain the confirmed and tentative variables of a Boruta object it has to be ensured that it has the variables Confirmed and Tentative. These can be obtained via the variables switch of the feature.boruta.comp method that has to be TRUE. Then the features contained in both variables will be combined into a Selected variable. Finally the method binds the Selected variable to the .GlobalEnv as features_boruta.

Value

Boruta object with an attached Selected variable containing the confirmed and tentative features.

Examples

 boruta <- KaggleHouse:::feature.boruta(
   target = data_train_na$SalePrice, predictors = data_train_na[-81],
    fixNA = T, roughFix = F, variables = T, verbose = T
 )
 KaggleHouse:::feature.boruta.selected(boruta)

MarcoNiemann/kaggle_house documentation built on May 7, 2019, 2:50 p.m.