Addressing Misinformation in Online Social Networks: Diverse Platforms and the Potential of Multiagent Trust Modeling

In this paper, we explore how various social networking platforms currently support the spread of misinformation. We then examine the potential of a few specific multiagent trust modeling algorithms from artificial intelligence, towards detecting that misinformation. Our investigation reveals that specific requirements of each environment may require distinct solutions for the processing. This then leads to a higher-level proposal for the actions to be taken in order to judge trustworthiness. Our final reflection concerns what information should be provided to users, once there are suspected misleading posts. Our aim is to enlighten both the organizations that host social networking and the users of those platforms, and to promote steps forward for more pro-social behaviour in these environments. As a look to the future and the growing need to address this vital topic, we reflect as well on two related topics of possible interest: the case of older adult users and the potential to track misinformation through dedicated data science studies, of particular use for healthcare.