I've occasionally been asked, especially since I decided to go "all in" by getting another graduate degree,1 what data science has to do with archaeology.2 This will be the first of a short series of articles to demonstrate how I've been using data, statistics, and such to gain insights into what is a surprisingly convoluted area of research. Basically, while many data scientists are trying to make inferences about the present or future I've been finding ways to apply those same methods to understanding the past.
Although I'm using an archaeological example, the methods presented here can be used for any number of research questions in other fields. I'll be covering some basics of:
-
Part I -- Creating and exploring bipartite and one-mode graphs [code]
-
Part II -- Similarity measures for sets and graph adjacency [code]
-
Part III -- Graph structure and community detection methods,
-
Part IV -- Geo-spatial networks
I'll be using R for the coding, but all of this could be done in Python just as easily.
Footnotes
-
Georgia Institute of Technology Master of Science in Analytics from the College of Computing, Scheller College of Business, and College of Egineering. I highly recommend the program! ↩
-
I'm also sometimes asked the inverse -- what does archaeology have to do with data science -- by my archaeological colleagues as well. Quantitative and computational (or "digital") archaeology has been around for a long while, but we're still something of a minority. That's a subject for a different post, though. ↩