Plotting a Network Graph using network_g2g

2025-06-17

This script demonstrates how to generate and visualize a network graph using the network_g2g function. The function creates an igraph object, which can be plotted to display connections between organizations based on collaboration metrics.

Step 1: load libraries and sample data

In this example, we will use the sample g2g_data dataset from the vivainsights package. We will also use dplyr and purrr for data manipulation and iteration respectively, as well as the igraph package for network graph creation and visualization.

library(vivainsights)
library(dplyr)
library(igraph)
library(ggplot2)
library(ggraph)
library(RColorBrewer)

# Display the first few rows of the dataset
head(g2g_data)

## # A tibble: 6 × 11
##   PrimaryCollaborator_Organization PrimaryCollaborator_…¹ SecondaryCollaborato…²
##   <chr>                                             <dbl> <chr>                 
## 1 Sales and Marketing                                 137 HR                    
## 2 Sales and Marketing                                 137 Unclassified Collabor…
## 3 Sales and Marketing                                 137 CEO                   
## 4 Sales and Marketing                                 137 Within Group          
## 5 Sales and Marketing                                 137 Product               
## 6 Sales and Marketing                                 137 Finance               
## # ℹ abbreviated names: ¹​PrimaryCollaborator_GroupSize,
## #   ²​SecondaryCollaborator_Organization
## # ℹ 8 more variables: SecondaryCollaborator_GroupSize <dbl>, MetricDate <chr>,
## #   Percent_Group_collaboration_time_invested <dbl>,
## #   Group_collaboration_time_invested <dbl>, Group_email_sent_count <dbl>,
## #   Group_email_time_invested <dbl>, Group_meeting_count <dbl>,
## #   Group_meeting_time_invested <dbl>

Step 2: Generate the igraph network object

The network_g2g() function constructs a network graph based on collaboration data. We set:

• primary and secondary to define the connection points
• metric to specify the weight of relationships
• return = "network" to get an igraph object

g <- network_g2g(
    data = g2g_data,
    primary = "PrimaryCollaborator_Organization",
    secondary = "SecondaryCollaborator_Organization",
    metric = "Group_meeting_time_invested", 
    return = "network"
)

Step 3: Prepare and customize the graph for visualization

Before plotting, we refine the graph by:

• Removing loops (self-connections) and multiple edges (redundant links)
• Setting vertex sizes based on the “org_size” attribute
• Defining a layout using Multidimensional Scaling (MDS)

# Simplify the graph (remove redundant edges and self-loops)
g <- simplify(g, remove.multiple = TRUE, remove.loops = TRUE)

# Scale node size based on organizational size
V(g)$size <- V(g)$org_size  

# Generate the MDS layout for better visual clarity
layout_mds <- layout_with_mds(g)

Step 4: Customize and plot the network graph

We set vertex and edge properties for better readability:

• Vertex color → blue
• Edge color → grey
• Vertex labels → organization names
• Edge width → fixed at 2 for clarity
• Transparency (alpha) for edges (adjusted manually)

# Example: Assign colors based on an attribute (assuming 'group' exists in V(g))
unique_groups <- unique(V(g)$group)  
color_palette <- rainbow(length(unique_groups))  # Generate distinct colors

# Assign colors based on group
vertex_color <- color_palette[as.numeric(factor(V(g)$group))]

# Use a color gradient from the 'Blues' palette
vertex_color <- colorRampPalette(brewer.pal(9, "Blues"))(length(V(g)))

# Assign colors based on scaled size values
vertex_color <- vertex_color[rank(V(g)$size)]

set.seed(123)  # For reproducibility
vertex_color <- sample(colors(), length(V(g)), replace = TRUE)

# Define vertex and edge attributes
#vertex_color <- "blue"  # Set all vertices to blue
edge_color <- "grey"    # Set all edges to grey
vertex_labels <- V(g)$name  # Use vertex names as labels

# Plot the graph with customized settings
plot(
  g,
  layout = layout_mds,           # Use the MDS layout
  vertex.color = vertex_color,   # Set vertex colors
  vertex.label = vertex_labels,  # Assign labels
  vertex.frame.color = "black",  # Define frame color for vertices
  vertex.size = V(g)$size,       # Scale vertex size
  edge.color = edge_color,       # Set edge colors
  edge.width = 2,                # Define edge width
  edge.alpha = 0.5,              # Adjust transparency (workaround)
  vertex.label.dist = 4          # Control label positioning
)

Step 5: Visualizing the Network Graph using ggplot2

Instead of using the base plot() function, we can leverage ggplot2 along with ggraph for a more refined and customizable visualization.

# Convert the igraph object into a dataframe for plotting
edges_df <- as_data_frame(g, what = "edges")
vertices_df <- as_data_frame(g, what = "vertices")

# Generate Layout (MDS)
layout_mds <- layout_with_mds(g)
vertices_df$x <- layout_mds[, 1]
vertices_df$y <- layout_mds[, 2]

vertices_df$color <- viridis::viridis(n = nrow(vertices_df), option = "C")

ggraph(g, layout = "mds") +
  geom_edge_link(aes(edge_alpha = 0.5), color = "grey", width = 2) +  
  geom_node_point(aes(size = size, color = size)) +
  geom_node_text(aes(label = name), vjust = 2, size = 2) +  
  scale_color_viridis_c() +
  theme_minimal() +
  labs(title = "Collaboration Network Graph", subtitle = "Using ggplot2 and ggraph") +
  theme(panel.grid = element_blank(), 
        axis.title = element_blank(), 
        axis.text = element_blank(), 
        axis.ticks = element_blank())

This final plot displays a network of collaborations based on the meeting time invested between different organizations.