Geofacts: A geo-reliability tool to empower fact-checking

In this paper, we present Geofacts, a web-based tool designed to support and enhance fact-checking services and help them to contrast the spread of fake news. Geofacts aims to facilitate the estimation of news reliability both for computer services and for human consumers by providing them with an ad hoc output. This tool uses the geographical characteristics extracted from the text and metadata present in a news to carry out a sequence of cross-checks between various typical elements of a news, such as the supposed place of the news, the area of action of the authors, the geographical origin of the IP address of the server, and the georeferencing components of the photos to calculate a geographical reliability index (GRI). This index ranges from 0 to 6, where 6 indicates a high level of reliability between the geographical features and the text content of the news, while 0 indicates no reliability or non-applicable process. Indeed, Geofacts was designed to work with news in which the geographic component is relevant, for example, for urban chronicles of episodes that happened in specific locations. On the other hand, there are contexts where the geographical component is not relevant or absent, for example, when the news is related to general football league news. Finally, Geofacts was tested on a set of news from three different specific topics from a popular news website. The results, although on a small set of topics, showed a high reliability of the algorithms used for the design and showed the Geofacts potential of supporting the contrast to the spread of fake news over the internet.