The impresso project draws on the expertise and talent of four leading institutions in digital humanities, computational linguistics and digital history from Luxembourg and Switzerland.


Luxembourg Center for Contemporary and Digital History, University of Luxembourg.

Founded in 2016,the C²DH focuses on the study of digital history, contemporary European and Luxembourgish history and public history. The centre promotes interdisciplinary research with a particular emphasis on developing and critically reflecting on new digital methods and tools, and embraces digital methods in its research, teaching and publications (e.g. the Journal of Digital History). C²DH has unique expertise in participatory design and development of methodologically reflected research software like impresso, histograph and DHARPA as well as widespread asset management tools (Zotero, Tropy).

C²DH has accumulated extensive experience in the design and development of methodologically reflected research software for the exploration and analysis of historical data. Within the project, C²DH is responsible for the participatory design and development of user interfaces, several case studies in digital history research and the coordination of dissemination activities.

Digital Humanities Laboratory, EPFL.

Founded in 2012, the DHLAB focuses on computational approaches for processing, accessing and studying large cultural heritage objects. Its research explores a variety of fields including linguistics, literature, history, art history and architecture, supported by methods from computer science, computer vision,natural language processing, geographic information science and web development. The DHLAB develops methods and softwares for information extraction and mining of historical texts, for transcribing and processing ancient documents and for reconstructing and visualising geographical spaces.

The DHLAB contributes expertise in the acquisition of sources (data sharing agreements, copyright management and data transfer), the development of text mining techniques (especially NE and image processing), the design, implementation and deployment of the integrated technical framework (data management, modelling and infrastructure), the organisation of shared tasks and computational work on historical case studies.

History department, University of Lausanne, UNIL.

The History Department has extensive expertise in media history and is actively engaged in research and teaching activities in digital history. In 2006, a research and teaching centre for the integration of audiovisual sources in contemporary history was set up, and collaborations with RTS, SRG SSR and Memoriav were also developed. Since 2011, the Department has been examining the contribution of digital technology to historical research (cf. transperiod research hub on digital history).

UNIL contributes its considerable expertise and networks in media – especially radio – and digital history, carries out several historical case studies, organises international conferences and helps strengthen the links with partner institutions, especially RTS and Memoriav.

Department of Computational Linguistics, UZH.

The department has been the leading text technology institution in the German-speaking part of Switzerland for the past 20 years. Its members are experts in machine learning for NLP and have contributed to neural approaches for machine translation, multilingual morphology, normalisation of historical spelling and OCR post-correction. Its activities include interdisciplinary text mining in historical, biomedical, political, social and linguistic projects, and the group has successfully completed projects to build, annotate and query large corpora for various languages. Another focus is cultural heritage engagement by combining human and machine reading for OCR crowd-correction and building a digital letter network of the reformer Heinrich Bullinger.

UZH contributes its machine learning expertise in text technology for annotating, indexing, aligning and exploiting large multilingual heritage corpora. In particular, it develops transfer-based methods for cross-lingual semantic analysis at word, phrase, paragraph and document level.