Network centrality graph showing the Twitter users most central to the conversation and the interaction between them. The graph consist of two elements, nodes and edges.
Nodes represents Twitter users in a community. The larger the node the more influencial the user (see the 3 ranking options under Node Rank). Select a node to see more information about that user. You can also drag and drop a node to more easily uncover its relationships with other nodes.
Edges between nodes indicate direct communications between the users. The thicker the edge the more communication between the two users. Edges can be shown as undirectional or directional, the latter will show the direction of a communication.
Graph Weight is a relative number from 1 to 25 where 25 is strongest. It indicates how influential this node based on who is interacting with this node within the graph you are seeing.
Graph Edges is the actual count of links between this node and other nodes within the graph you are seeing.
First section of the toolbar allow you to Zoom Out, Zoom Reset, and Zoom In. You may also zoom in and out with gestures on touchpad/touch screen, or with mouse wheel.
Node and Edge Controls
The second section of the toolbar offers a varierty of node and edge controls:
Node Rank. Choose between 3 algorithms for how to rank nodes.
- Mentions. A simple arithmetic rank based on number of mentions.
- Authority. An advanced SNA algorithm ranking nodes based on authority within community.
- Hub. An advanced SNA algorithm ranking nodes based on its function as a hub within a community.
This slider allows you to slim the graph by removing the nodes with lesser weight.
Node Colors. Controls the color of the nodes.
- Gradient Blue. This is the default color scale, darker blue indicates higher node weight.
- Healthcare Stakeholder. Nodes will be colored by Healthcare Stakeholder. Legend will explain colors.
- Random Gradient 1-color. For every selection of this option, a random 1-color scale is produced. Repeat selection to find a color scale that suits your need.
- Random Gradient 2-colors. Color scale between two colors.
- Random Gradient 3-colors. Color scale between three colors.
Node Lables. Controls the node labels.
- Switch freely between Screen Name and Display Name.
- Increase or reduce the number of nodes with labels with the slider.
Edges. Controls the edge style.
- Simple. Unidirectional edges. Dashed. Dotted Line.
- Directional. Arrow. Curved Arrow. Tapered.
- Toggle All. Select all or none of the Healthcare Stakeholders.
- Colorize Nodes. Nodes will be colored by Healthcare Stakeholder. Legend will explain colors.
- Check or uncheck single categories to show or hide those within that category.
Select Detect Communities to run an algorithm that will detect sub-communities within the graph. The algorithm will identity and colorize those nodes that it finds to be closer to eachother in the graph.
Learn more about the Louvain community detection algorithm.
Download Node and Edge Data.
- Download as Image
- Export Nodes to CSV
- Export Edges to CSV
- Export to Gephi (only available to Enterprise users)
Use the share button to the far right in the toolbar to share the graph with the same filtering, color and layout with another user.
- Limit. Sets the maximum number of nodes to be shown.
- The Node Rank algorithms Authority and Hub are by nature reductive in that they will reduce the number of Nodes in the graph. To counter this, you may opt to increase the graph size by increasing the Limit setting before using the two algorithms.