The NewsEye project aimed to improve access to early European press (1850–1950) for researchers, library users, and the public. Using 15 million digitised pages from the national libraries of Austria, Finland, and France, it developed automatic tools for character recognition, newspaper structure analysis (themes, article identification), and multilingual content processing (recognition of mentions of people, places and organisations, opinion analysis, text mining). The result is a toolkit for large-scale analysis of digitised newspapers in multiple languages and from diverse sources.
The consortium included three national libraries and eight academic partners: four in the humanities and social sciences, and four in computer science (Universities of La Rochelle, Helsinki, Innsbruck, Rostock, Paul-Valéry Montpellier, and Vienna). The project was primarily funded by the EU’s Horizon 2020 research and innovation programme.
NewsEye introduced new methods for content analysis and exploration tailored to historical documents. It generated new insights in history, literature, gender studies, and media analysis across French, German, Finnish, and Swedish corpora—demonstrating its value across the social sciences and humanities, regardless of the language of the sources.
The project renewed the state of the art in document analysis and natural language processing, achieving numerous international benchmarks. It enabled a deeper understanding of textual content despite imperfect digitisation and OCR, paving the way for progress in automatic analysis of historical documents—an area where large language models still lag. In recognition of its impact, NewsEye received the European Heritage Europa Nostra Award 2024 in the research category.