City and Google working to combat fake news

City, University of London received a €335,000 grant from the Google Digital News Initiative (DNI) fund in July 2017 to develop a web-based tool called ‘DMINR’.

The DNI is a partnership between Google and publishers in Europe to support high-potential ideas to benefit the news ecosystem, in areas such as automation, verification and news storytelling.

Academics from the Department of Journalism and the Centre for Human-Computer Interaction Design will spend two years working on the DMINR application to help journalists make connections in data, verify information and combat the issue of fake news.

"Separating the news from the noise is key to the verification of digital information. DMINR serves to empower journalists by supporting them in identifying and making sense of connections in large, complex datasets."

Tom Felle, Senior Lecturer in Digital Journalism at City, who leads the DMINR project.

Tom Felle leads the DMINR project.

The tool will utilise lateral search and application programme interface (API) aggregation to find connections in big data and multiple lateral search locations. It will incorporate a highly advanced search capability, artificial intelligence mark-up language (AIML) and a visual connection finder interface to then make sense of, verify and visualize those connections. It will be designed and evaluated from a user-centred perspective to ensure it delivers a high-quality user experience.

“The easiest way to think about what the API aggregator does is via the use of mobile apps that allow us to check bus timetables or flight information. These apps use timetables published by transport companies, as well as real-time information, to give a result,” said Felle. “We will then build an algorithm to find the most useful and relevant answers, similar to how Google finds what it thinks are the top answers to the questions one might type into a search bar.”

The DMINR team, which also includes Dr Glenda Cooper (Lecturer in Journalism), Dr Stephann Makri (Senior Lecturer in Human-Computer Interaction) and Dr Andrew MacFarlane (Reader in Information Retrieval), will collaborate with data teams from news organisations including The Guardian and The Telegraph to test the application.