AI-Driven Analysis and Mapping of NATO S&T Data

The NATO Science & Technology Organization (STO) plays a critical role in advancing the Alliance's strategic posture by aligning science and technology (S&T) investments with defence and security priorities. A cornerstone of this effort is the ARTEMIS¹database, developed under the auspices of the NATO Science & Technology Board (STB). ARTEMIS records correlated S&T activities in NATO², S&T taxonomies/ontologies, and political-military impact areas to enable inter alia portfolio management, inform decision-making and support capability development. Central to this paper are advanced methodologies to address labor-intensive biased manual mapping. Using semantic similarity approaches powered by embedding models and vector databases, ARTEMIS now employs automated multi-text classification and clustering mechanisms to identify synergies, reduce duplication, and map S&T activities to impact areas and taxonomies with precision. These innovations enhance operational impact by providing the Nations and NATO with a robust platform for strategic-, planning-, and tactical-level decision-making. This paper was originally presented at the NATO Science and Technology Organization Symposium (ICMCIS) organized by the Information Systems Technology (IST) Scientific and Technical Committee, IST -209-RSY - the ICMCIS, held in Oeiras, Portugal, 13–14 May 2025.