(under review at EMNLP’22) Characterizing Harmful Agendas in News Articles
Abstract: Manipulated news online is a growing problem which necessitates the use of automated systems to curtail its spread. However, these systems must be interpretable given the sensitivity of related issues like censorship. Automatically characterizing news articles requires detection of the article’s factuality, any authorial deception, and its agenda. We argue that while misinformation and disinformation detection have been studied, there has been a lack of investment in the important open challenge of detecting harmful agendas in news articles; identifying harmful agendas is critical in order to flag the most insidious manipulated news campaigns online. In this work, we propose this new task and release an initial dataset, NewsAgendas, of annotated news articles for agenda identification. We show how interpretable systems can be effective on this task and demonstrate that they can perform comparably to black-box models.