Detecting influential news in online communities: An approach based on hexagons of opposition generated by three-way decisions and probabilistic rough sets

The work proposes a new method to detect influential news in online communities. Influential news are articles that induce shifts in users’ opinions or, in general, lead to a polarization of opinions or change like-mindedness of users. The method aims at supporting online platform managers and editors in understanding the impact that social content and news can have on the dynamics of opinions. The influential news detection is conduced by using the Three-Way Decisions approach based on Probabilistic Rough Sets to perform a tri-partitioning of online users. The three parts are then mapped onto a structure, namely Hexagons of Opposition, allowing to reason on opinions, related to a given set of news, of specific communities. More in detail, several hexagons of opposition are constructed along the timeline, as recent news are considered, and compared to detect which news contribute to change opinions of the considered communities. Moreover, two indicators have been introduced to measure the impact of the news. The proposed method has been experimented on real data derived from existing datasets and the promising results have been discussed by using a qualitative approach.