An effective response to humanitarian crises relies on detailed information about the needs of the affected population.

Current approaches to assessing humanitarian needs through surveys often require interviewers to convert complex, open-ended responses into simplified categorical data. More nuanced insights require the use of qualitative methods, but proper transcription and manual coding are hard to conduct rapidly and at scale during a crisis. As a result, the amount and usefulness of qualitative information to inform humanitarian assistance are severely limited.

Natural language processing (NLP), a form of artificial intelligence, provides potentially far-reaching new opportunities to capture qualitative data from voice responses and analyze it for relevant content to better inform humanitarian assistance decisions.

This project, launched in 2018, consists of two main activities:

1. Design a pilot system using NLP to transcribe, translate, and analyze large sets of qualitative responses to a population-based humanitarian need assessment survey with a view to improving the quality and effectiveness of humanitarian assistance.
2. Anticipate the ethical challenges of introducing this new technology and other automated decision systems to the humanitarian context— and create a framework to reduce and mitigate these new risks.

Background Literature

Needs Assessment Handbook, United Nations High Commissioner for Refugees
Humanitarian Needs Assessment: The Good Enough Guide, ACAPS
Needs Assessment and Analysis, United Nations Office for the Coordination of Humanitarian Affairs

Project Team

Tino Kreutzer, PhD Candidate, School of Health Policy and Management
James Orbinski, Director, Dahdaleh Institute for Global Health Research
Lora Appel, Postdoctoral Research Fellow, OpenLab
Aijun An, Professor, Department of Electrical Engineering and Computer Science, York University
Muath Alzghool, Postdoctoral Fellow, York University

You may also be interested in...