Anthology of Computers and the Humanities · Volume 4

Les embeddings , nouvel outil d’une histoire numérique des grands corpus de presse?

Arthur Michelet1 and Martin Grandjean1

  • 1 Université de Lausanne

Permanent Link: https://doi.org/10.63744/R2LLVRsBeNuI

Published: 21 May 2025

Keywords: History, economic history, media history, public relations, natural language processing, embeddings, digital humanities, computational humanities, data visualization

Mots clés : Histoire, histoire économique, histoire des médias, relations publiques, traitement automatique des langues, embeddings, humanités numériques, humanités computationnelles, visualisation de données

Abstract

This research focuses on the development of a procedure for analyzing very large corpora of news articles. It discusses the cross-analysis of archival sources and news articles using text embeddings within a notebook designed for historians and students.