This workshop will explore how networks can be incorporated into research in the humanities. We will discuss the basics of network analysis, finding and preparing network datasets, types and methods of measurements, and techniques for creating attractive visualizations.
Participants will learn how to use Gephi, an open source program for network visualization and analysis, and the workshop will conclude with a planning activity for possible future research involving networks.
Bio: Chris Alen Sula
Chris Alen Sula is Assistant Professor at the School of Information & Library Science at Pratt Institute and Coordinator of Digital Humanities. He received his PhD in Philosophy from the City University of New York with a doctoral certificate in Interactive Technology and Pedagogy. His research focuses on network studies of intellectual and social movements, as well as digital humanities and information visualization.
I’m working with Gephi for analyzing semantic links among textual fragments. Would love to learn more!
Happy to hear what people are working on / interested in! Let us know on this thread.
Sounds Great! I’m curious to learn more about this software and its possible applications for research.
Hey, Chris — probably no surprise to you that I’m very interested in this, both for my dissertation and for the Writing Studies Tree. Amanda and I have played around with Gephi, but only briefly and with little success so far (though that was back in its alpha release, and I see there’s a beta out now). Looking forward to your guidance!
Are there any materials you’d advise us to bring? e.g. a preferred data sample size or format that we can follow along with?
Great question, Ben. I think I’ll start by focusing on some more background issues–how to find/create data, what kinds of things can be represented in a network, etc. This is probably useful as a standalone topic (for people who just drop in), and it often makes me rethink my datasets.
If you want to follow along with Gephi, it would help to have an Excel or csv file with two columns of data showing links from one thing to another (e.g.,
Ben Miller, CUNY Graduate Center
Chris Sula, Pratt Institute
…
Regarding size, I’ve used Gephi up to 30K points (nodes) and 100K connections (edges), but it sometimes crawls at that scale (though it should be able to handle 1M edges). Something on the order of hundreds or a thousand edges usually works nicely, and I’ll provide a link to whatever data I use in my walkthrough in case people want to follow along.