feature.boruta.formula: Create Formula Based on Boruta Selected Features

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

Description

The method creates a formula that puts a target value in relationship to features selected by the Boruta algorithm.

Usage

feature.boruta.formula(boruta, tentative = F)

Arguments

boruta	Boruta object obtained by the execution of feature.boruta method.
tentative	boolean switch to decide whether tentative variables should be part of the created formula. Default is FALSE.

Details

This method evaluates the Boruta object. In case some other object is provided to the function it will abort with an error. Given correct inputs the function accesses the target property of the custom Boruta object returned by feature.boruta and also obtains the categorized list of the evaluated variables via feature.boruta.variables. With these inputs a formula of the structure target ~ predictorVar1 + ... is created which can then e.g. be reused for regression.

Value

formula object that represents the relationship between the predictors and the target variable which has been evaluated by the Boruta algorithm. The form will be target ~ predictorVars where predictorVars will only be confirmed (and tentative) variables.

Examples

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

 KaggleHouse:::feature.boruta.formula(boruta, tentative = T)

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