The News Angler project aims to support journalists finding new and unexpected angles on unfolding news stories, along with suitable background information. The project therefore explores how artificial intelligence (AI) techniques can leverage big open data sources to support high-quality journalism. Central AI techniques so far are knowledge graphs and natural-language processing.
Our evolving News Hunter platform harvests potentially news-related information in real time from textual sources, such as social media, commercial news aggregators, and open reference sources. Knowledge graphs and related semantic techniques are used to represent and integrate this diverse information as networks that can be stored, analysed, and shared more easily, precisely, and in new ways.
The project is a collaboration between the University of Bergen (UiB) and Wolftech Broadcast Solutions AS, a developer of newsroom tools for the international market. The cross-disciplinary project team at UiB belongs to the research group for Intelligent Information Systems (I2S), located in the Department of Information Science and Media Studies and associated with Media City Bergen. The project employs a Post-Doc and a Ph.D. and involves several other collaborators and guests.
The project continues a long-standing research collaboration between Wolftech and I2S Group that started already in 2015, when Wolftech saw a need for more effective tools for journalists – the current tools on the market did not pay sufficient attention to practical needs.
Year one summary
A central task in the first year of the project was conducting and preparing publication of systematic literature reviews. The research group did technical groundwork and planned the doctoral and post-doctoral projects. The group has presented and published several contributions to international and national conferences. Themes have been architecture for the News Angler platform, the concepts of news angles and newsworthiness from an ICT perspective, analogical reasoning about news angles, and ontologies for representing journalistic information in network form. The first journal articles from the project are now being evaluated by renowned international journals.
Year two summary
In its second year, the project has published several papers in international journals and made many conference and workshop presentations. Several other papers are in preparation or under review. In addition to presenting research results, we have prepared several comprehensive literature reviews. The News Hunter platform is up and running, after having been reimplemented from scratch using state-of-the-art big data and cloud technologies. Harvesters, basic lifters, and central data stores are in place, and work continues on more advanced text analysis and lifting and on representing and reasoning over news angles.
Year three summary
In its third year, the project has continued to publish research results in international journals and at conferences and workshops. The News Hunter platform has been extended with new components for analysis and lifting of text, relation extraction, lifting and aggregation of GDELT data, and detecting location angles. We have also held a two-day meeting with our advisory board. We are working to create synergy between News Angler and the new MediaFutures Research Centre for research-driven innovation, where Wolftech is a partner and the research group behing News Angler leads a work package. We are also planning further projects in the continuation of News Angler.
Sources: Andreas L Opdahl, Arne Berven, and Ingrid Aarseth Johannessen,