create.eval.cart: Build & Evaluate Classificaion & Regression Trees (CART)

In guptadeepak18/DTAME: Build & Evaluate Decision Trees

Description

This function takes a dataframe as an input along with the name of the response variable. It then sets the seed as per user or default value and split the data into training and testing sets. After the split it builds a CART model on training data and evaluates the performance of the model on different Evaluation methods. It can also optimize the CART tree by pruning it as per the least Cross Validation error, user has to define if they want the model to be pruned or not.

Usage

create.eval.cart(x, response, seed = 42, splitRatio = 0.7,
  minbucket = 10, cp = 0, prune = FALSE, plot = TRUE)

Arguments

x	Data Frame that you want to work with
response	Dependent(Target) Variable
seed	Seed Value, default = 42
splitRatio	Ratio to split the dataframe into Train and Test
minbucket	Minimum number of observations that should be present in each leaf/terminal node
cp	Cost Complexity Paramter Value
prune	Pruning Tree on the basis of minimum Cross Validation Error (xerror), default = FALSE
plot	Plot the model Tree and CP Table

Value

A List with Model Attributes & Evaluation Results

Author(s)

Gupta, Deepak [e-mail]

Examples

create.eval.cart(sleep, group, plot = FALSE)

guptadeepak18/DTAME documentation built on Dec. 25, 2019, 9:14 a.m.