Presentation title:
Automated Multilingual Text Analysis in the Europe Media Monitor
Presentation description:
The European Commission’s 'Joint Research Centre' (JRC) has developed the family of 'Europe Media Monitor' (EMM) applications that help information-seeking users find and analyse live information in their field of interest. By automatically pre-processing the textual input, EMM supports the users in their intellectual analysis of large amounts of textual data: EMM gathers and automatically analyses a daily average of 220,000 online news articles in seventy languages, plus selected social media posts; EMM groups articles on the same events; For each article, it recognises and disambiguates names of persons, organisations and locations; It recognises quotations by and about people; EMM offers automatic translations into English; And it hyperlinks related articles in other languages. An important feature for the human analyst are trend visualisations that guide the analysis and highlight peaks and exceptional trends: How certain news stories have developed over time; Which news categories are most active at the moment; What combinations of country and threat currently have the highest alert levels world-wide. EMM was developed for the EU institutions and for national authorities of all EU Member States, but it is also used by international organisations (e.g. United Nations sub-organisations) and by EU partner countries around the world. The public can access EMM freely via web interfaces and mobile Apps. EMM uses over 1000 categories from a wide range of subjects, with a focus on Global Security and Crisis Management. The speaker will present EMM functionality and give an insight into how the output is used in real-life scenarios. He will give concrete examples of news content complementarity across languages (national bias).