Network analysis methods are frequently - and increasingly - applied to collections of data about historical people, from the ancient period through to the modern. The utility of these methods remains debated, however; findings that appear interesting or significant can be readily challenged by pointing out the non-representativeness of the dataset that was analysed.

The purpose of this project is to carry out a meta-analysis of several collections of historical network data, ranging from the medieval period to the modern, and use them to explore how the usual network analysis methods are impacted by the shapes of the respective datasets. Participants will be free to select networks to analyse, in consultation with the supervisor, and will choose the methods of analysis. It is anticipated that the project results will be published in a suitable academic journal after the semester ends.

Prerequisites for students: Python programming; Familiarity with, or willingness to learn, basic statistical analysis; Familiarity with, or willingness to learn, network analysis methods

Project open to: Business Analytics, Data Science, Digital Humanities

Number of students: 2-4