How to Customize CTREE Decision Tree Plot in R

Suppose you want to change a look of default decision tree generated by CTREE function in the party package in R.

Decision Tree Custom Ctree Plot

Create Sample Data

Let's create a sample data for demonstration purpose. Here we created a binary variable for using it as a target variable in the decision tree.

library(dplyr)
iris2 <- iris %>%
  mutate(Target = as.factor(ifelse(Species == "setosa", "Setosa", "Not Setosa"))) %>% 
  select(-Species)

The ctree function is used to create the tree and then the plot function is used with custom panel functions to customize the look of the tree.

library(party)
t1.prior <- ctree(Target ~ ., data = iris2)
plot(t1.prior, type="simple", 
     inner_panel=node_inner(t1.prior,abbreviate = FALSE, pval = TRUE, id = FALSE), 
     terminal_panel=node_terminal(t1.prior, abbreviate = FALSE, digits = 2, fill = c("white"),id = FALSE)
)

How does the above code work?

1. ctree(Target ~ ., data = iris2): This line creates a conditional inference tree model with the binary target variable Target and all other variables in the iris2 dataset as independent variables.
2. plot(...): This is used to plot the created tree with the following customizations.
   • type="simple": Specifies the type of plot, which is a simple plot.
   • inner_panel=...: Specifies the custom panel for inner nodes using the node_inner function. The abbreviate argument is set to FALSE to display full variable names, pval = TRUE displays p-values, and id = FALSE hides the node IDs.
   • terminal_panel=...: Specifies the custom panel for terminal nodes using the node_terminal function. The abbreviate argument is set to FALSE to display full variable names, digits = 2 sets the number of digits displayed, fill = c("white") sets the fill color for terminal nodes, and id = FALSE hides the node IDs.

Related Posts

R Tutorials : Top 100 R Tutorials

Spread the Word!
Share Share Tweet

About Author:

Deepanshu founded ListenData with a simple objective - Make analytics easy to understand and follow. He has over 10 years of experience in data science. During his tenure, he worked with global clients in various domains like Banking, Insurance, Private Equity, Telecom and HR.

While I love having friends who agree, I only learn from those who don't
Let's Get Connected Email LinkedIn