Predicting Controversial News Using Facebook Reactions

Different events and their re- ception in different reader communities may give rise to controversy. We pro- pose a distant supervised entropy-based model that uses Facebook reactions as proxies for predicting news controversy. We prove the validity of this approach by running within- and across-source exper- iments, where different news sources are conceived to approximately correspond to different reader communities. Contextu- ally, we also present and share an au- tomatically generated corpus for contro- versy prediction in Italian