Fake News and its Credibility Evaluation by Dynamic Relational Networks: A Bottom up Approach

The world has become flat through the Web and the recent proliferation of social network services, allowing anyone to be a news writer without requiring much effort, thus creating an epidemic of fake news (microblogs in particular). This note proposes a bottom up approach with relative, mutual, and dynamic credibility evaluation using a dynamic relational network (or mutual evaluation model), where each node can evaluate and in turn be evaluated by other nodes for credibility based on the consistency of the content of the node. Our stance is that 1) it would be difficult to evaluate credibility of news solely on the content of the news (except obvious ones using fake photos and images); 2) hence, allow related news articles to mutually evaluate each other based on the facts’ (5W1H— who, what, where, when, why, and how) consistency; 3) however, this poses the problem of fake evaluation, thus allowing any users to build their own evaluation network for fairness.