Abstract: 

In an initial emergency response, having the right information in the right hands is key to saving lives. When that emergency is cascading, it's critical for responders to have up-to-date information in order to make decisions that can impact people’s safety as a situation unfolds. In both singular and cascading scenarios responders and peacekeepers alike must be equipped to quickly make decisions that may include what resources to deploy and where. Fortunately, in most emergencies, today people lean on social media to publicly share information and at the same time sensor data is increasingly becoming more publicly available. But, a platform developed to detect emergency situations while also delivering the right information has to be capable of ingesting thousands of noisy data points per second: sifting through and identifying relevant information, from different sources, in different formats, with varying levels of detail, in real-time, so that relevant individuals and teams can be alerted at the right level and at the right time.

In this talk I will describe the technical challenges in processing vast amounts of heterogeneous, noisy data in real-time from the web and other sources, highlighting the importance of interdisciplinary research and a human-centered approach to address problems in humanitarian and emergency response. I will give specific examples and discuss relevant future research directions in several AI fields.

Bio: 

Aoife Cahill is a Natural Language Processing (NLP) expert and a director of AI research at Dataminr, the leading real-time information discovery platform. Since joining in 2021, Aoife has led a team of data scientists focused on the efficient iterative process of developing and evaluating AI technology that supports the expansion of Dataminr’s internal and external products.

Prior to Dataminr, Aoife led a team of research scientists and engineers working on high-stakes NLP applications in the educational domain at the Educational Testing Service (ETS). The NLP teams at ETS are known leaders in the field of developing and deploying robust, well-documented, scalable NLP prototypes that maintain fairness across user groups.

Aoife holds a PhD in Computational Linguistics from Dublin City University, Ireland, and has also spent time conducting NLP research in Germany, Norway and in the U.S. As an active member of the computational linguistics research community, her research has been published in top-tier journals including Computational Linguistics and the Journal of Research on Language and Computation, as well as conference proceedings at the annual conference for the Association for Computational Linguistics (ACL), the International Conference on Computational Linguistics (COLING) and the Conference on Empirical Methods in Natural Language Processing (EMNLP).