COMRADES project witnessed its final review at the European Commission in Brussels recently, bringing to an end the work of the project which started in January 2016. COMRADES is a €2 million European H2020 project, co-ordinated by Prof Harith Alani from KMi, and involves the University of Sheffield (UK), Delft University of Technology (The Netherlands), Ushahidi network (Kenya), and Gov2U (Belgium).
COMRADES (Collective Platform for Community Resilience and Social Innovation during Crises) aims to empower communities with intelligent socio-technical solutions to help them reconnect, respond to, and recover from crisis situations. The project’s core innovations are built around the collection and processing of information related to crisis events, to raise people’s awareness through social-enabled platforms.
In the final review report, the project officer and reviewers acknowledged and praised the exceptional quality of the work and achievements of the project over its lifetime of three years. They wrote that “Excellency has been reached” and that “the project has largely overpassed the expected impact and exploitation plan”. In an acknowledgement of the role of Prof Harith Alani as the project Co-ordinator, the reviewers stated that the project’s success “would not have been possible without efficient and effective management during the 36 months of the project.”
Over the past three years, COMRADES produced several technologies, such as:
- Ushahidi COMRADES platform: allows the gathering of information, incidents, and reports from people directly or via Twitter. The primary source code for the functionality of the platform is available on Github. Also available is the source code that fetches data from the services developed for COMRADES, such as YODIE, CREES, and EMINA.
- CREES (Crisis Event Extraction Service): an advanced Artificial Intelligent method that identifies crisis events in social media streams. CREES is available as an integrated component in the COMRADES platform, but also as an API and as a Google Sheet plugin. CREES was designed and developed by Gregoire Burel; a Research Associate in KMi.
- EMINA: and informativeness classifier for social media posts, to filter the deluge of data on social media and to selects which messages are more informative and which are more actionable.
- YODIE: an antity finder for social media posts. The tool automatically identifies uncommon entities in given posts, to help digital responders and others to very quickly learn about the locations, organisations, and other entities mentioned in the post.
- VERACITY: a veracity classifier for social media messages, which automatically calculate the probable veracity level of a given message.
The reviewers stated that “the use of these tools would most probably reduce the loss of life as well as environmental, material and economic losses.” In particular, they praised the use of the COMRADES technologies to monitor Kenya’s General Elections, and said that that “was a good opportunity to test the COMRADES platform because it was a good chance to see how the platform and all the tools included here work in a real-time situation.” In May 2018, COMRADES was featured at the European Commission portal as a success story.
COMRADES produced over 30 peer reviewed publications, targeting multiple fields and disciplines, such as Humanitarian Technology, Computer Science, Human-Computer Interaction, and Crisis Management. It has also produced 27 deliverables.