GGPlot Flavored Treemaps

Treemapify

There are several options for visualizing treemaps in R. This note focuses on Treemapify, a package maintained by David Wilkins.

I favor this approach over the treemap package because it is compatible with ggplot2 and allows users to access its’ functionality.

Here’s an example Treemap I created to visualize the dominant emotions displayed for the iconic 90’s sitcom, Friends. You can find out more about the Friends dataset here.

Other visualizations I created for the Friends project can also be found here.

Below, we can see that geom_treemap, geom_treemap_subgroup_border and geom_treemap_subgroup_text are layers that works seamlessly with other layers like scale_fill_manual, theme, and labs that are staples of the ggplot2 package.

Bottom line, it’s easier to customize treemaps from the treemapify package than the treemap package.

library(treemapify)

ggplot(friends_emo_tree, aes(area = n, label = speaker, subgroup = emotion)) +
    geom_treemap(aes(fill = emotion, alpha = n)) +
    geom_treemap_subgroup_border(color = 'white') +
    geom_treemap_subgroup_text(place = 'bottom', grow = T, alpha = 0.3, color = 'black',
                               min.size = 0) +
    geom_treemap_text(color = 'white', fontface = 'italic', place = 'centre', reflow = T) +
    scale_fill_manual(values = c('#FF4238', '#FFDC00', '#42A2D6', '#9A0006', '#FFF580', '#00009E')) +
    theme(
        plot.background = element_rect(fill = '#36454F'),
        legend.position = 'none',
        title = element_text(colour = 'white', family = 'Friends')
    ) +
    labs(
        title = 'The One with the Dominant Emotions',
        caption = 'Viz: @paulapivat | Data: #TidyTuesday'
    )