Modeling Controversy within Populations

A growing body of research focuses on computationally detecting controversial topics and understanding the stances people hold on them. Yet gaps remain in our theoretical and practical understanding of how to define controversy, how it manifests, and how to measure it. Since controversy is a complicated social phenomenon, it is difficult to understand what elements make up the controversy. Previous work has attempted to capture controversy algorithmically by studying cues for disagreement and polarity between different stance groups. However, we still lack a systematic understanding of how controversy should be defined and measured. In this paper, we propose a multi-dimensional model of controversy. Specifically, we introduce a model with two minimal dimensions: contention and importance. Our model departs from existing work by viewing controversy as a trait rooted in population. It suggests that controversy should be separately observed in a given population, rather than held as a fixed universal quantity. We model contention and importance within a population from a mathematical standpoint. To validate and evaluate the soundness of our theoretical model, we instantiate the model to algorithms for a diverse set of sources: polling, Twitter, and Wikipedia. We demonstrate that our controversy model holds an explanatory power for observed phenomena but also a predictive power for tasks, such as identifying controversial Wikipedia articles.