Interactive Visualization of Ancient Social Networks

Interactive visualizations of graph data can be valuable tools to understand relationships across a variety of domains. But how do you convert 4000 year old letters into a format that can be graphed? And then, how can you graph this data and share it with people?

How do I process my data so that it "looks" like a graph?

Lots of data can be viewed as a graph, but right now, it probably looks more like a spreadsheet.  One of the most challenging parts of visualizing any data can be putting it into a format that can be read and understood by a graphing program.

Using Python, we were able to write simple scripts to convert an exported comma-separated file into Node and Edge tables that can be loaded into a variety of graph visualization frameworks.

How do I actually use a visualization tool?

One popular Graph Visualization tool is called Gephi.  It is open-source and cross platform, meaning that anyone can easily use it for free.  Using Gephi, it is possible to align nodes using various algorithms, interactively manipulate nodes, and compute various graph statistics.  Due to the large size of many graphs, being able to filter graphs down to smaller viewable sections is an important property, and Gephi can be used to view the nodes involved in a certain subset of letters, or involving only a single person.  We were able to learn how to use this software, and share these insights to help make manageable visualizations.

How can I share my research?

We were able to produce an interactive graph that can be used online using the Sigma.js javascript library  This can be embedded online and shared with other researchers, so that your work can be easily built upon by other researchers.

Get in touch!

E-mail us any time to set up a free consulting session: support@help.hmdc.harvard.edu