PropEval – The Automatic Detection of Propaganda Techniques in News Articles

As part of SemEval-2020, I built two machine learning classifiers that attempted to detect and label propagandistic rhetorical techniques as used in American news articles. I was honored to be part of an initiative that pushed the boundaries of current technologies in computational pragmatics, as well as to work on something that could be beneficial to the standard of quality and honesty of information available in American news media. Certainly not least, undertaking a project of this scope was an intense and hugely beneficial learning experience.

View The Project on Github Here

Download the Paper (8-page version) Here

Download The Presentation Here