Title: The assessment of Information Credibility in the Social Web

Abstract: In the scenario of the Social Web, where a large amount of User Generated Content is diffused through Social Media without any form of trusted external control, the risk of running into misinformation is not negligible. For this reason, assessing the credibility of “potential” information constitutes a fundamental issue. Credibility is a quality perceived by individuals, who are not always able to discern with their own cognitive capabilities genuine information from fake one. For this reason, in the last years several approaches have been proposed to automatically assess credibility of UCG in Social Media. Most of them are based on data-driven approaches, based on machine learning techniques, but recently also model-driven approaches are emerging.

Model-driven approaches aim at defining a predictive model based on an analysis of the problem and of the identified objects and their features; in particular, approaches relying on a Multi Criteria Decision Making paradigm constitute a way to compute an overall credibility assessment associated with a given information (posts and blogs) by separately evaluating each feature connected to each alternative, and by subsequently aggregating the single assessments into a global one. Several classes of aggregation operators can be employed to obtain the overall credibility estimate, thus modeling distinct behaviors of the considered process. Furthermore, some aggregation operators allow model the interaction between criteria. In this lecture an overview of model driven approaches will be presented, and their application to some real problems will be illustrated.