The COMRADES (www.comrades-project.eu/) consortium came back triumphant from Brussels where the project’s mid-term review was held at the European Commission on July 5th. Project officer and reviewers were highly impressed by the exceptional quality of the work and achievements of the project partners over the first half of the project. 

In these 18 months, the consortium developed several advanced methods for crisis information processing, including a semantic deep-learning approach that beats the state of the art in automatically grouping social media posts with respect to the type of crisis-related information they contain, and a novel entity extraction tool that identifies and disambiguates unknown place names from social media content. They also praised the extensive community engagement activities carried out by the project, which involved hundreds of stakeholders from several countries around the world, including Nepal and China. 

Reviewers also had the chance to use the first prototype of the COMRADES platform, which offers the functionalities of the Ushahidi (www.ushahidi.com/) crisis-mapping tool, with added A.I. services for powerful processing of disaster information. 

COMRADES is a €2 million European H2020 project, co-ordinated by Prof Harith Alani from the Open University, and involves the University of Sheffield (UK), Delft University of Technology (The Netherlands), iHub (Kenya), and Gov2U (Belgium).