Circular Packing to visualise hierarchy data in R

In this article, we are talking about handling hierarchical data using circular packing visualizations. To prepare circular packing with R Programming Language, we will use ggraph package and prepare a bubble to show the hierarchies. 

Circular Packing to visualize hierarchy data in R

Preparing the Hierarchical Data

Here we are going to prepare hierarchical data for demonstration. For this, we will use flare datasets.

R

# Libraries
library(ggraph) # to prepare visualisation
library(igraph) # for network analysis
library(tidyverse) # for data handling
library(viridis) # for generating the color map

# data for hierarchical structure
edges = flare$edges
head(edges)

Output:

Creating another dataframe for hierarchical structure

R

vertices = flare$vertices
head(vertices)

Output:

Preparing the graph with dataframe:

R

# preparing the graph
mygraph <- graph_from_data_frame( edges,
                                  vertices = vertices )
mygraph

Output:

IGRAPH 6e05b59 DN-- 252 251 --  
+ attr: name (v/c), size (v/n), shortName (v/c) 
+ edges from 6e05b59 (vertex names): 
[1] flare.analytics.cluster->flare.analytics.cluster.AgglomerativeCluster 
[2] flare.analytics.cluster->flare.analytics.cluster.CommunityStructure   
[3] flare.analytics.cluster->flare.analytics.cluster.HierarchicalCluster  
[4] flare.analytics.cluster->flare.analytics.cluster.MergeEdge            
[5] flare.analytics.graph  ->flare.analytics.graph.BetweennessCentrality  
[6] flare.analytics.graph  ->flare.analytics.graph.LinkDistance           
[7] flare.analytics.graph  ->flare.analytics.graph.MaxFlowMinCut          
[8] flare.analytics.graph  ->flare.analytics.graph.ShortestPaths          
+ ... omitted several edges

Visualize Circular Hierarchy 

Here we will visualize the dataframe with a hierarchical structure.

R

# plot the graph using ggraph
ggraph(mygraph, # graph data
       layout = 'circlepack',

       # size of bubbles based on
       # the size parameter in vertices data
       weight = size) +
  geom_node_circle(aes(fill = as.factor(depth),
                       color = as.factor(depth) )) +
  # define the color of each different labels
  scale_color_manual( values=c("0" = "green", "1" = "red",
                               "2" = "red",
                               "3" = "red", "4"="red") ) +
  scale_fill_manual(values = c("0" = "green", "1" = viridis(4)[1],
                               "2" = viridis(4)[2], "3" = viridis(4)[3],
                               "4" = viridis(4)[4])) +
  theme_void()

Output: