In his talk at the Digital Media Research Center of the Queensland University of Technology (QUT), Fabio Giglietto of the University of Urbino explored three distinct methods he and his research team employed using OpenAI models to identify the most salient topics circulated via Facebook links in the run-up to the two most recent Italian general elections.
He furthermore elaborates on techniques for constructing a classifier to detect political links shared on Facebook, performing a cluster analysis on the document embeddings provided by OpenAI’s API embedding endpoint for political links, and autonomously labeling the identified clusters.
Finally, Fabio outlined utilising Meta’s URL Shares Dataset to characterise each cluster based on their exposure and interaction patterns.
Author: Jochen Spangenberg (DW)