Balanced News Using Constrained Bandit-based Personalization
This addresses the issue of filter bubbles and polarization in news consumption for users, though it appears incremental as it applies existing constrained bandit methods to a new domain.
The researchers tackled the problem of polarized news feeds by developing a prototype news search engine that presents balanced viewpoints across liberal and conservative articles, using constrained bandit optimization to allow flexible user-defined constraints and showcasing it side-by-side with a traditional polarized feed.
We present a prototype for a news search engine that presents balanced viewpoints across liberal and conservative articles with the goal of de-polarizing content and allowing users to escape their filter bubble. The balancing is done according to flexible user-defined constraints, and leverages recent advances in constrained bandit optimization. We showcase our balanced news feed by displaying it side-by-side with the news feed produced by a traditional (polarized) feed.