Developing AI predictive migration tools to enhance humanitarian support

The EUMigraTool (EMT) provides short-term and mid-term predictions of asylum seekers arriving in the European Union, drawing on multiple sources of public information and with a focus on human rights. After 3 years of development, it has been tested in real environments by 17 NGOs working with migra...

Descripción completa

Detalles Bibliográficos
Autores: Blasi Casagran, Cristina|||0000-0002-4327-2212, Stavropoulos, Georgios
Tipo de recurso: artículo
Fecha de publicación:2024
País:España
Institución:Universitat Autònoma de Barcelona
Repositorio:Dipòsit Digital de Documents de la UAB
Idioma:inglés
OAI Identifier:oai:ddd.uab.cat:304361
Acceso en línea:https://ddd.uab.cat/record/304361
https://dx.doi.org/urn:doi:10.1017/dap.2024.76
Access Level:acceso abierto
Palabra clave:Predictive tools
Migration
Humanitarian tools
EUMigraTool
Artificial intelligence
Descripción
Sumario:The EUMigraTool (EMT) provides short-term and mid-term predictions of asylum seekers arriving in the European Union, drawing on multiple sources of public information and with a focus on human rights. After 3 years of development, it has been tested in real environments by 17 NGOs working with migrants in Spain, Italy, and Greece. This paper will first describe the functionalities, models, and features of the EMT. It will then analyze the main challenges and limitations of developing a tool for non-profit organizations, focusing on issues such as (1) the validation process and accuracy, and (2) the main ethical concerns, including the challenging exploitation plan when the main target group are NGOs. The overall purpose of this paper is to share the results and lessons learned from the creation of the EMT, and to reflect on the main elements that need to be considered when developing a predictive tool for assisting NGOs in the field of migration.