About this Site

This site explores--and offers users the opportunity to explore--the rhetoric of computational propaganda that occurred on Facebook during the 2016 election. The project was developed by Dr. Damien Smith Pfister, Nora Murphy, Meridith Styer, and Misti Yang in collaboration with Purdom Lindblad and Ed Summers at the Maryland Institute for Technology in the Humanities. 160 students from the "Interpreting Strategic Discourse" classes offered by the University of Maryland's Department of Communication coded the dataset by hand.

The IRAdS website contains over 3,000 Facebook advertisements that the Internet Research Agency, a Russia-linked “troll farm,” purchased in the run-up to the 2016 election campaign. This is one of the most sophisticated efforts at computational propaganda yet, but little systematic analysis has been done on this data corpus.

Pfister, Murphy, and Yang developed the codebook based on concepts developed in the class (e.g. metaphors, myths, ideographs, semiotics; syllabus available here). Our hope is that this dataset will surface and organize different themes across these advertisements. In collaboration with MITH, these advertisements will be posted, with our analysis embedded as metadata, on a website that other publics can use to better understand Russia’s propaganda efforts.

Soon, we will run some preliminary analysis of this data, which students will draw on to support their own analysis of the advertisements for their final paper.

Data Sources

This database contains metadata and images that were extracted from the Social Media Advertisements: data set released by Facebook to the House Intelligence Committee, who then released them to the public. The original PDF data was processed by extracting the post images and metadata so that it could be loaded into this site.