Plot a Treee Object

This function visualizes either the entire decision tree or a specific node within the tree. The tree is displayed as an interactive network of nodes and edges, while individual nodes are scatter/density plots using ggplot2.

Usage

# S3 method for class 'Treee'
plot(x, datX, response, node = -1, ...)

Arguments

x

A fitted model object of class Treee, typically the result of the Treee() function.

datX

A data frame of predictor variables. Required for plotting individual nodes.

response

A vector of response values. Required for plotting individual nodes.

node

An integer specifying the node to plot. If node = -1, the entire tree is plotted. Default is -1.

...

Additional arguments passed to the plotting functions.

Value

A visNetwork interactive plot of the decision tree if node = -1, or a ggplot2 object if a specific node is plotted.

Overall Tree Structure

A full tree diagram is displayed using visNetwork when node is not specified (the default is -1). The color represents the most common (plurality) class within each node, and the size of each terminal node reflects its relative sample size. Below each node, the fraction of correctly predicted training samples and the total sample size for that node are shown, along with the node index. Clicking on a node opens an information panel with additional details.

Individual Node Plot

To plot a specific node, you must provide the node index along with the original training predictors (datX) and responses (response). A scatter plot is generated if more than one discriminant score is available, otherwise, a density plot is created. Samples are projected onto their linear discriminant score(s).

Examples

fit <- Treee(datX = iris[, -5], response = iris[, 5], verbose = FALSE)
plot(fit) # plot the overall tree

plot(fit, datX = iris, response = iris[, 5], node = 1) # plot a specific node