A 2-hour hands-on workshop. Build research collaboration networks from sponsored research data using Gephi, and learn the fundamental concepts and statistical metrics behind them.
Get the materials →A short overview of what we'll build together.
Be able to build a research collaboration network from sponsored research data.
Introduce some fundamental concepts around networks and statistical metrics.
The post-it wall analogy for research networks. Why sponsored research data works. A walkthrough of the Gephi interface.
Load data, build a simple network, add names, run statistics, apply filters.
Degree, betweenness centrality, closeness centrality, modularity. Guided learning through an AI chatbot tailored to your role.
Automatic research group detection using modularity. Coloring the network, sizing by degree, using the Data Laboratory.
Move from Overview to Preview, tune export settings, produce a searchable PDF. A note on dynamic (time-based) networks.
Review the underlying metrics table in the Data Laboratory.
Use your own data, or use the provided FY21 and FY21–FY23 federal proposal datasets. Compare and contrast the two.

Awareness-level coverage. You'll recognize these terms and what they mean when you see them in Gephi.
How many connections a person has. PI → Co-I is a directed network, so in-degree and out-degree differ.
How often someone sits on the shortest path between two other people. Identifies connectors between groups.
Who is "closest" to everyone else. For each person, average the number of edges it takes to reach every other node.
Analyzes all connections to determine groups. In our case, research clusters. We color nodes by modularity class to surface research teams.
The shared folder contains the slide deck, Gephi, dummy proposal data, and interactive AI chatbots. Install Gephi before the workshop.
Slide deck, Gephi installer, dummy proposal data, and the interactive AI chatbots.
Google Drive → PDF · SlidesView the complete session deck directly in your browser.
Open PDF → Tool · ExternalOpen-source desktop application to visually analyze networks. Windows, Mac, and Linux.
gephi.org → Prior workIntegrating data analytics for enhanced university outcomes. Background reading on the approach behind this workshop.
Substack →All thoughts shared in this workshop are the presenter's own and do not necessarily represent those of any affiliated institution.