Libya Humanitarian Monitoring Dashboard
How might we improve humanitarian needs assessment and monitoring by collecting and analyzing public data streams?
The Office for the Coordination of Humanitarian Affairs in the Region of the Middle East and North Africa (OCHA ROMENA) is responsible for planning and monitoring humanitarian aid for the 2.44 million affected people in Libya. Because of the instability of the country, UN agencies and NGOs have very limited data about humanitarian aid needs in the country. As a result, plans are built leveraging the sparse data they have, which is often incomplete and anecdotal.
Unite Labs initiated this project in partnership with OCHA ROMENA, Microsoft, and Global Pulse to examine the potential of analysis of publicly available data - Twitter, Facebook, and popular news sites -to help OCHA ROMENA improve their ability to assess the needs of people in Libya.
After several months of background research, including an OCHA-led fact-finding mission in Tunis, the technological groundwork was created during a four-day sprint in New York. During this time, the OCHA Libya team was embedded with and co-created the early prototype alongside the development team.
Incoming data is filtered, analyzed with Natural Language Processing, and run through a machine learning service that is learning to infer location and conversations related to key humanitarian clusters - including Early Recovery, Food Security, Health, Protection, and Water, Sanitation, and Hygiene.
All of this will be delivered to the team as a visual dashboard that can be filtered by category, time, and location.
When deployed, this will help the Libya team identify and respond to needs more proactively, share more data back to key agencies and NGOs for better response monitoring, create a stronger evidence base for decision-making, and more easily build the case for future resources.
During the prototyping stage, the objective is to build, test, and improve a dashboard that provides the Libya Humanitarian Country Team with data-driven insight on the the current needs for humanitarian aid and help to create more accurate and effective plans to respond to these challenges. The team is exploring and learning about opportunities in sentiment analysis and predictive analytics.
Next, the project will move into the pilot stage. During this stage the Libya team will run the technology in parallel with its standard practices, continuing to train the model and report back on its accuracy. This research will guide and inform this project’s ability to scale, and also inform the team about possible future opportunities in public information and media monitoring technologies.
The technology itself will be released under an open source license. A long-term objective for this project is to create a shareable resource for other humanitarian groups and UN agencies that can be adapted, learned from, and reused across variety of regional contexts and related challenge areas (for example, using a modified version of this technology to help the Department of Political Affairs anticipate armed conflict).