Tag Archives: artificial intelligence

Kaspar Gubler: Von der Sammlung Stelling-Michaud zum Repertorium Academicum Helveticum (1938-2026)

Abbilung: Wirkungsräume von Basler Studenten 1460-1550, beschriftete Orte mit 60 oder mehr Ereignissen zu Tätigkeiten, mehrheitlich im kirchlichen und städtischen Bereich, Datengrundlage: repac.ch, 05/26

Vortrag von Kaspar Gubler im Rahmen der internationalen Fachtagung des Forschungsprojekts Repertorium Academicum (REPAC): Die Schweiz und die europäischen Universitäten im Mittelalter Grundlagen für die moderne Wissensgesellschaft, 28. / 29. Mai 2026, Universität Bern.

Der Vortrag behandelt den Weg von der Karteikartensammlung Sven Stelling-Michauds zum Repertorium Academicum Helveticum als digitalem Forschungsinstrument. Im Mittelpunkt stehen Wiederentdeckung, Digitalisierung und wissenschaftliche Integration eines lange verschollen geglaubten Quellenkorpus. An diesem Beispiel wird gezeigt, wie analoge Vorarbeiten in eine digitale, erweiterbare Forschungsumgebung überführt werden können und wie dadurch neue Perspektiven auf Mobilität, Wissensräume und Universitätsgeschichte der Schweiz im Spätmittelalter entstehen. Damit erweitert sich zugleich der Deutungsraum prosopographischer Forschung: Historische Personen, Orte, Institutionen und Wissensbestände werden nicht nur gesammelt, sondern räumlich, zeitlich und sozial miteinander in Beziehung gesetzt. KI gestützte Verfahren können diesen Deutungsraum künftig weiter öffnen, sofern sie quellenkritisch kontrolliert und in Forschungsumgebungen eingebunden werden.

Programm REPAC Workshop(PDF)

Data, Dialogue and Discovery @ Bern Data Science Day (8 May 2026): Poster on AI-workflows in historical research

At this event, Kaspar Gubler will present a poster on AI-supported workflows for historical research.

The poster presents a controlled workflow for AI-supported historical research developed in the context of the Repertorium Academicum (REPAC) project at the Historical Institute, University of Bern. Using nodegoat as a research environment, the workflow connects structured prosopographical data on European academics between 1250 and 1550 with AI-supported methods for information retrieval, data enrichment, and analysis.

Two complementary workflows are highlighted. The first focuses on source processing and data enrichment, including translation of historical texts, named entity recognition, text tagging, and structured extraction into nodegoat. The second combines curated REPAC data with large language models through a retrieval-augmented generation pipeline: relevant data are retrieved first and then supplied as contextual grounding for dialogic analysis.

The poster argues that AI can support discovery in historical research when embedded in transparent, data-driven environments. At the same time, it reflects critically on possible effects of accelerated AI-assisted workflows, including reduced depth of subject engagement and fewer serendipitous discoveries. The central principle is: retrieve first, interpret second.

Bern_Data_Science_Day_2026_nodegoat (PDF)

Kaspar Gubler: Tracing Manuscript Circulation and Exile Networks through Digital Methods: The Example of Johannes (Jan) Škréta (ca. 1600 – 1651)

Fig. Basel, Universitätsbibliothek, AN II 21: Matricula facultatis medicae II, 1570-1814 (https://www.e-codices.unifr.ch/de/list/one/ubb/AN-II-0021), Matriculation at the University of Basel April 1620 (no 986):  Johannes Škréta Schotnovius a Zavorzitz, Pragensis, Bohemus

Program_Manuscript_Prague_2026 (PDF)

“This presentation introduces a reusable data model developed within the nodegoat environment, designed to map the complex networks of early modern exiles. While the model is illustrated through a case study of exiles in Switzerland (1621–1638), specifically focusing on the figure of Johannes Škréta, its architecture is intentionally generalizable for application across different regions and time periods. The core of the data model relies on a tripartite structure of Persons, Institutions, and Manuscripts. Crucially, the manuscript component extends beyond simple bibliographic data to capture the dynamic lifecycle of the object, including ownership history (purchase and sale transactions) and physical circulation. Furthermore, this presentation will demonstrate how AI methodologies can be integrated with nodegoat to enhance text analysis and the study of manuscript provenance. To ensure practical utility for the research community, the full data model will be made freely available. I will conclude with a practical demonstration showing how the Prague nodegoat community can import this schema into their own infrastructure and adapt it to their specific research questions.”

Sample data model for Manuscripts, Persons, Institutions (for Import into nodegoat):

nodegoat_manuscript_model_gubler.json (download)

See the tutorial how to import this model into your nodegoat installation.

Short version of the tutorial:

1. Create an API client in Management in nodegoat
2. Assign a user to the API client (this generates the pass key)
3. Download the sample model from this website, unzipp it
4. Open your command-line tool (for instance, Terminal on macOS)
5. Put this into your command-line tool:
curl -H “Authorization: Bearer N4zU7eX8TnC00N9D20boEAUkhSqecd8TLmbbaz5kK3123456” https://api.nodegoat.yourdomain.com/model/type -X PUT -d @
Replace the pass key and domain
7. Drag the model file (.json) after …/type -X PUT -d @
In my case: …@/Users/kaspar/Desktop/nodegoat_manuscript_model_gubler.json and press Enter; the import starts
8. Create a project in nodegoat and activate all Manuscript types and classifications

To make the visualisations work with the model, you need to configure the visual settings according to nodegoat principles (scope):

https://nodegoat.net/documentation.s/81/scope

 

Conference organised by the project:

Manuscript Practices and the Textuality of Exile Communities from the Czech Lands in the 1620s and 1630s

CFP Manuscript Practices and the Making of Exile Communities in the Early Modern Period

https://www.hsozkult.de/event/id/event-157335

For recent research on Johannes Škréta see:

Holý, Martin, unter Mitwirkung von Kamil Boldan, Vojtěch Pelc, Ondřej Podavka, Marie Ryantová und Marta Vaculínová. Die Universität Basel und die Böhmischen Länder (1460–1630). Ostfildern: Thorbecke Verlag, 2025, PDF: https://shop.verlagsgruppe-patmos.de/media/medien/pdf/ebook/9783799521253_ebook.pdf

https://shop.verlagsgruppe-patmos.de/die-universitaet-basel-und-die-boehmischen-laender-1460-1630-402045.html

 

Kaspar Gubler: The Data Iceberg: Context and Interpretation in AI-Supported Digital History

Fig. Vector-based analysis within the research environment nodegoat.

Presentation by Kaspar Gubler within the workshop: Researching with AI. Between Opportunity and Imposition, Workshop, University Bern, February 11, 2026

Abstract:

Artificial Intelligence offers powerful new opportunities for Digital History, particularly in the analysis of large, structured datasets. At the same time, it risks obscuring what historical data fundamentally consists of: uncertainty, context dependence, and interpretation. This post introduces an iceberg model that distinguishes between visible, formalised data (names, places, events) and the largely invisible layers of historical meaning beneath the surface. Using the prosopographical research database REPAC as an example, it shows how AI-supported methods, especially Retrieval-Augmented Generation (RAG) in nodegoat, can help to address uncertainty without flattening it. AI thus becomes not a substitute for historical interpretation, but a tool that more tightly connects data, context, and research questions.

Gubler_The-Data-Iceberg_Context-and-Interpretation-in-AI-Supported-Digital-History (PDF)

Program_AI_Workshop_University_Bern_2026 (PDF)

Artificial Intelligence and the Study of University Knowledge Spaces in the Holy Roman Empire in the Middle Ages

Presentation at the Atelier Heloise workshop by Kaspar Gubler, Pim van Bree, and Geert Kessels, November 13-14 November 2025, Prague.

This presentation explores how generative AI and large language models can enhance the study of university knowledge spaces in the Holy Roman Empire during the Middle Ages. Using prosopographical data from the Repertorium Academicum (REPAC) within the nodegoat environment, it illustrates how these tools can reveal hidden structures of academic mobility and intellectual exchange. The talk also addresses the epistemological and hermeneutic implications of applying generative models to historical data.

Atelier_Heloise_Program_Abstracts_Prague_2025 (PDF)