Network Analysis

The most common social media platforms that are easy to scrape are Twitter and Reddit. Increasingly, scholarship is also focusing on Youtube, Facebook and non traditional spaces, such as ArtStation, Dribbble, DeviantArt), academy (ResearchGate, Academia) and music (Spotify, Soundcloud).

These are some examples of projects utilizing social media web scraping at the Scholars Studio:

Before diving into historical network analysis, you have to begin with a dataset pulled together from various historical sources. This source material will differ based upon a given project. Below you will find some basic starting points for collecting historical network data, with an emphasis on the ancient and medieval world. Keep in mind that this list is not exhaustive:​

1. Epigraphic Sources: Stone inscriptions including the names of individuals from the Greco-Roman world
   • Corpus Inscriptionum Latinarum/CIL (Latin inscriptions): http://cil.bbaw.de/cil_en/index_en.html​
   • Epigraphic Database Heidelberg (Greek and Latin inscriptions): http://edh-www.adw.uni-heidelberg.de/home
2. Prosopographies: Lists of individuals from various societes, based upon numerous sources
   • Roman World (260-641 CE)–Jones, A. H. M., J. R. Martindale, and John Morris, eds. The Prosopography of the Later Roman Empire, Volumes I-III. Cambridge: Cambridge University Press, 1971-1992.
   • Byzantine World (1025-1305 CE)–Prosopography of the Byzantine World: http://blog.pbw.cch.kcl.ac.uk/
   • Anglo-Saxon England (6th-11th c. CE)–Prosopography of Anglo-Saxon England: http://www.pase.ac.uk/index.html
3. Other Digitized Sources: 
   • ​​HathiTrust: https://www.hathitrust.org/
   • Internet Archive: https://archive.org/index.php

Although there are several different types of network analysis tools, each one requires datasets that contain digitized information pulled from source materials. After locating your source materials, the next step will be to create a spreadsheet containing all of the relevant information needed to create a network. Again, this is entirely dependent on which tool you use and what type of network you are interested in creating. For example, to create simple network showing a historical family, a spreadsheet would need to include, at least, the following:

1) Individual's Name     2) Individual's Gender     3) Associated Family Members