Create interactive 3D network visualization widget

Creates an interactive 3D visualization of your Bluesky social network using WebGL. The visualization shows users as nodes colored by community membership and connections as edges, allowing you to explore network structure, identify clusters, and understand relationship patterns in your social network.

Usage

create_widget(net, profiles, prop = 1)

Arguments

net

Tibble. Network edge list with follow relationships containing:

actor_handle

Character. User who is following another

follows_handle

Character. User being followed

profiles

Tibble. User profile information from get_profiles containing handle, displayName, description and other profile details

prop

Numeric. Proportion of users to include in visualization (0-1). Use smaller values for large networks to improve performance (default 1.0)

Value

An interactive HTML widget (threejs object) that displays:

• Nodes representing users, colored by community

• Edges representing follow relationships

• Interactive controls for rotation, zoom, and exploration

• Hover tooltips showing user information

Details

The visualization process includes:

1. Community detection using random walk algorithm

2. Color assignment based on community membership

3. Node labeling with user display names and descriptions

4. 3D layout generation for optimal viewing

For large networks (1000+ users), consider using prop < 1 to sample a subset for better performance. The widget works best with networks of 50-500 users for detailed exploration.

See also

build_network, add_metrics, trim_net

Examples

if (FALSE) { # \dontrun{
# Build a network for visualization
auth <- get_token("your.handle.bsky.social", "your-app-password")
network <- build_network(
  key_actors = c("scientist.bsky.social", "researcher.bsky.social"),
  keywords = c("science", "research", "academic"),
  token = auth$accessJwt,
  refresh_tok = auth$refreshJwt,
  threshold = 15,
  prop = 0.3  # Sample 30% for manageable size
)

# Create full network visualization
full_widget <- create_widget(network$net, network$profiles, prop = 1.0)

# Display the interactive visualization
full_widget

# Create smaller sample for performance
sample_widget <- create_widget(network$net, network$profiles, prop = 0.5)
sample_widget

# For very large networks, use even smaller samples
if (nrow(network$profiles) > 500) {
  performance_widget <- create_widget(network$net, network$profiles, prop = 0.2)
} else {
  performance_widget <- create_widget(network$net, network$profiles, prop = 0.8)
}

# Save widget as HTML file for sharing
if (requireNamespace("htmlwidgets", quietly = TRUE)) {
  htmlwidgets::saveWidget(full_widget, "network_visualization.html")
  message("Visualization saved as network_visualization.html")
}

# Customize for specific communities
# Filter to show only the largest communities
profiles_with_metrics <- add_metrics(network$profiles, network$net)
large_communities <- profiles_with_metrics[
  profiles_with_metrics$community %in% c(1, 2, 3),  # Top 3 communities
]
filtered_net <- network$net[
  network$net$actor_handle %in% large_communities$handle &
  network$net$follows_handle %in% large_communities$handle,
]
focused_widget <- create_widget(filtered_net, large_communities, prop = 1.0)
focused_widget
} # }