Data Visualization with Treemaps in R

Introduction

Treemaps are a powerful visualization tool used to display hierarchical data using nested rectangles. Each rectangle represents a category, with the size and color of the rectangles representing additional variables.

Importance of Treemaps

Visualizing Hierarchies: They effectively display hierarchical structures, showing the relationship between parent and child categories.

Comparing Proportions: They allow for easy comparison of the relative sizes of different categories.

Utilizing Space: They make efficient use of space, allowing a large amount of data to be displayed in a compact area.

Highlighting Patterns: They help identify patterns and trends within hierarchical data.

Creating Treemaps in R

To create treemaps in R, we will use the treemap package. Let’s start with a basic treemap and gradually explore more complex variations.

Basic Treemap

A basic treemap displays hierarchical data with rectangles sized according to a specified variable.

Code

library(treemap)

# Create a sample dataset
data <- data.frame(
  Category = c("A", "B", "C", "D", "E"),
  Value = c(30, 20, 25, 15, 10)
)

# Create a basic treemap
treemap(data,
        index = "Category",
        vSize = "Value",
        title = "Basic Treemap")

Treemap with Colors

Adding colors to a treemap can provide additional information about the categories.

Code

# Create a sample dataset with additional variables
data <- data.frame(
  Category = c("A", "B", "C", "D", "E"),
  Value = c(30, 20, 25, 15, 10),
  Group = c("Group 1", "Group 1", "Group 2", "Group 2", "Group 3")
)

# Create a treemap with colors
treemap(data,
        index = c("Group", "Category"),
        vSize = "Value",
        vColor = "Group",
        type = "categorical",
        title = "Treemap with Colors")

Treemap with Multiple Levels

Treemaps can display multiple levels of hierarchy, providing a detailed view of nested categories.

Code

# Create a sample dataset with multiple levels
data2 <- data.frame(
  Level1 = c("A", "A", "B", "B", "C"),
  Level2 = c("A1", "A2", "B1", "B2", "C1"),
  Value = c(30, 20, 25, 15, 10)
)

# Create a treemap with multiple levels
treemap(data2,
        index = c("Level1", "Level2"),
        vSize = "Value",
        title = "Treemap with Multiple Levels")

Treemap with Proportional Colors

Using proportional colors to represent an additional variable can provide deeper insights into the data.

Code

# Create a sample dataset with an additional variable
data <- data.frame(
  Category = c("A", "B", "C", "D", "E"),
  Value = c(30, 20, 25, 15, 10),
  Score = c(90, 80, 70, 60, 50)
)

# Create a treemap with proportional colors
treemap(data,
        index = "Category",
        vSize = "Value",
        vColor = "Score",
        type = "value",
        title = "Treemap with Proportional Colors",
        palette = "RdYlGn")

Using ggplot2

Code

library(tidyverse)
library(treemapify)
store <-  data.frame(Fruits = c("bananas","apple","melon","plums","pineapple",
                                "orange","apricot","grapes"),
  sales = c(25000,22000,15000,5000,18000,20000,3000,9000))
store %>% 
  ggplot(aes(area = sales, fill = Fruits, label = Fruits))+
  geom_treemap(layout = "squarified")+
  geom_treemap_text(place = "centre", size = 12)+theme(legend.position = "none")