Exploring the application of Large Language Models and other AI techniques to the European Crisis Management Laboratory analyses
The European Crisis Management Laboratory (ECML) supports policymakers in crisis preparedness and response, adding value to the context analysis, operational activities, and the related information products. At the same time, these analyses are meant to enhance situational awareness by identifying emerging dynamics and trends via anticipatory analyses.
The ECML performs scientific analyses for natural and man-made disasters, conflicts and complex crises at the global level via a daily monitoring by subject matter experts. Its reaction times can be minutes, when the impact assessment is performed via automatic systems like GDACS, the Global Disaster Alert and Coordination System, or a few weeks maximum. The European Crisis Management Laboratory analyses are provided in the form of (web) systems and services, scientific reports and maps, publicly distributed or for the exclusive use of the European policy Directorates Generals, depending on their sensitivity.
The use of Large Language Models (LLM) represents an important opportunity to enhance the efficiency and effectiveness of the scientific analyses performed by the ECML. Its adoption is more profitable if associated with techniques to collect information from selected, corporate large databases like those behind the ECML systems. This approach is called Retrieval Augmented Generation (RAG). It provides the users with texts drafted from a collection of selected, trusted sources of information that best fit the scope of the application. This way, not only the initial results of the AI-extracted information are more reliable for the analysts to use them, but also the AI engine is trained only with credible and reliable sources, thus becoming more and more trustworthy over time.
The ECML tested the possibility of including this functionality in the daily work of its disaster risk management analysts by integrating it with their information systems and analytical tools.
GALLIANO Daniele Alberto;
BITUSSI Alessandro;
CARAVAGGI I.;
DE GIROLAMO Ludovica;
DESTRO Destro;
DUTA Ana-Maria;
GIUSTOLISI Luca;
LENTINI Azzurra;
MASTRONUNZIO Marco;
PARIS Stefano;
PROIETTI Chiara;
SALVITTI Valerio;
SANTINI Marzia;
SPAGNOLO Luigi;
2024-12-15
Publications Office of the European Union
JRC138914
978-92-68-22801-2 (online),
1831-9424 (online),
EUR 40153,
OP KJ-01-24-204-EN-N (online),
https://publications.jrc.ec.europa.eu/repository/handle/JRC138914,
10.2760/0323818 (online),
