This data paper by Claude Grasland, professeur à l’Université de Paris Cité membre de Géographie-cités and Etienne Toureille presents a reproducible method for identifying national and supranational geographical entities mentioned in international news published by daily newspapers.
The approach relies on a multilingual dictionary covering macro-regions (part of the world and political organisations) and nation-states in four languages (French, German, English, Turkish). The dictionary has been tested and validated on a corpus of news titles from five daily newspapers in France, Tunisia, Germany, the United Kingdom, and Turkey, spanning the period from April 2013 to March 2023. This corpus serves as an example of the dictionary’s application for elaborating geographical networks and geopolitical agendas.
Claude Grasland, Etienne Toureille, « A multilingual dictionary for the recognition of macro-regions and countries mentioned in daily press news », Cybergeo: European Journal of Geography [En ligne], Data papers, document 1106, mis en ligne le 08 avril 2026 DOI : https://doi.org/10.4000/16cap
Procedure for the construction of the multilingual dictionary of macro-regional entities © Grasland and Toureille, 2026
Cybergeo
This article has been published in Cybergeo within a new form of peer-reviewed articles devoted to the description, documentation and evaluation of geographical databases produced by the authors. The article provides links to the data itself, which is stored on a sustainable repository and distributed under an open licence.
