Hi all, I recently finished up a project with some classmates of mine and I wanted to show it off. We took a dataset of thousands of conspiracy theory articles -- Miani et al's LOCO Corpus -- and went about analyzing it to find out how different conspiracy theories connect to each other.
I'd encourage everyone to take a look at the full article because there are some fun visualizations in there, but TLDR: We used topic modelling and sentiment analysis to find out what these articles are about and then keyword extraction and pattern mining to learn what conspiracy theories tend to occur together.
Show HN: Debunked Modelling the LOCO Conspiracy Corpus
logan-cooper.com