Opening up Echo Chambers via Optimal Content Recommendation

Online social platforms have become central in the political debate. In this context, existence of echo chambers is a problem primary relevance. These clusters like-minded individuals tend to reinforce prior beliefs, elicit animosity towards others and aggravate spread misinformation. We study phenomenon on Twitter dataset related 2017 French presidential elections propose method tackle it with content recommendations. use quadratic program find optimal recommendations that maximise diversity users are exposed to, while still accounting for their preferences. Our relies theoretical model can sufficiently describe how flows through platform. show provides good approximations empirical measures demonstrate effectiveness optimisation algorithm at mitigating chamber effect dataset, even limited budget