Finding Conflicts of Opinion in Citizen Participation Platforms

Online citizen participation platforms are powerful democratic tools that allow large numbers of contributors to be involved in public decision-making. However, for large groups of contributors to collaborate, we need to provide tools for users and decisionmakers to navigate and understand high volumes of content. Towards this goal, we introduce an approach based on natural language processing to detect pairs of contradictory and equivalent proposals in online citizen participation contexts. We apply this approach on two major national citizen consultations: the République Numérique and Revenu Universel d'Activité consultations. We highlight the potential of our method in two use cases. First, our method is a high-quality tool for finding idea communities in online content. Second, we demonstrate that the method improves on the state-of-the-art for finding relevant complementary content for a user, by identifying new relevant views for 76 % of the proposals tested.