Lab Topic

Collaborative Lab Notes Doc

Instructions

Hopefully, you’ve already downloaded Gephi, but if not you will need it for today’s workshop.

We will be using Melanie Walsh’s sample network datasets today as well as data from the Mapping the Republic of Letters project. I’ve gathered the files here for you to download (you will click the Download Raw File button, just under History). If you’re on a mac, double-click the zip to extract the files; if you’re on Windows you’ll need to right click and select Extract All.

There are many kinds of visualizations in digital humanities work—graphs, charts, maps, trees, timelines, &c.—but today we’ll focus on one: the network graph. Networks are used to analyze and visualize relationships. And though we often think of networks primary as visualizations, this lab will seek to demonstrate how those visualizations are only one aspect—and often not the most important aspect—of a network analysis.

Building on our discussion of Benjamin Schmidt’s argument in “Do Digital Humanists Need to Understand Algorithms?”, we will discuss a number of key metrics from the field of social network analysis—e.g. degree, centrality, betweenness—with an eye toward understanding how they are computed from network data.

Let’s start by looking together at this network graph of Star Wars characters. Note that if you hover over a node, the character name will appear. The creator has written about how this graph was created and what it represents, but I want you to first write down some hypotheses. What do you think is being measured about these characters here, and how is that measurement being determined?

Lab Task: Build & analyze your own network

Using either one of the datasets from Melanie Walsh that we did not discuss in class or assembling your own edges and nodes:

  1. Import or create an edges and nodes table
  2. Run the major network metrics: degree, centrality, betweenness, modularity. You can experiment with others.
  3. Experiment with the layout and visualization of your network based on these metrics. Be sure to look at the numbers in addition to the visualization itself.
  4. Draft a lab report, following the general guidelines for lab reports, that reflects on these qualities. What have you learned (or not) about your network using these methods? What would you need to better understand to take full advantage of these methods?

Resources

Updated: