Project Fortis: Accelerating UN humanitarian aid planning and response with GraphQL
How can humanitarian aid field experts use social media to gain insight, understand trends, and track key humanitarian issues? What can social media tell us that first hand data can’t? How can social media alert experts identify emerging trends that could have a profound impact? These are some of the questions that Microsoft and the United Nations set out to investigate and understand when we started working together on Project Fortis. What we learned was that by using advanced Machine Learning techniques, Apache Spark, GraphQL, React, integrated social media with a range of public and privately available data, and applying powerful data visualization techniques, we could indeed provide deeper insight and awareness to humanitarian disasters (i.e., refugee displacement, terrorist attacks, disease outbreaks, famine, etc.) Project Fortis was born. Project Fortis grew into a customizable tool that enables anyone to define a set of topics, a set of social media sources, and geography to track and monitor on their own. This talk focuses on the architecture of Fortis and how GraphQL was used to provide further insights into the post-gaddafi humanitarian situation in Libya.