Fact Checking Misinformation Using Recommendations From Emotional Pedagogical Agents

Dealing with complex and controversial topics like the spread of misinformation is a salient aspect of our lives. In this paper, we present initial work towards developing a recommendation system that uses crowd-sourced social argumentation with pedagogical agents to help combat misinformation. We model users' emotional associations on such topics and inform the pedagogical agents using a recommendation system based on both the users' emotional profiles and the semantic content from the argumentation graph. This approach can be utilized in either formal or informal learning settings, using threaded discussions or social networking virtual communities.