Thematic Landscapes of the Past: Analysing Slovene Historical Periodicals With Topic Modeling

Filip Dobranić, Uroš Šmajdek, Oliver Pejić, Ciril Bohak, Vojko Gorjanc, Tina Munda, Darja Fiser


Abstract
This paper explores the thematic landscapes of three Slovene historical periodicals—Slovenka, Slovenec, and Slovenski narod—from the sPeriodika corpus, a comprehensive collection of Slovene press published between 1771 and 1914. Using BERTopic, we analyse the thematic profiles of these periodicals, enriched with diachronic perspectives. Our study examines the thematic commonalities and specificities of the selected periodicals, highlighting their distinct political orientations, target audiences, and the increasing nationalist polarisation in public discourse. This work contributes to digital humanities by demonstrating the potential of modern topic modelling techniques, such as BERTopic, to advance historical and cultural research.
Anthology ID:
2026.pressmint-1.14
Volume:
Proceedings of the First Workshop on Creating Interoperable Corpora of Historical Newspapers
Month:
May
Year:
2026
Address:
Palma de Mallorca, Spain
Editors:
Maciej Ogrodniczuk, Petya Osenova, Tanja Wissik
Venues:
PressMint | WS
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
92–99
Language:
External URL:
https://lrec.elra.info/lrec2026-ws-pressmint-14
DOI:
10.63317/263a2ep7iiyv
Bibkey:
Cite (ACL):
Filip Dobranić, Uroš Šmajdek, Oliver Pejić, Ciril Bohak, Vojko Gorjanc, Tina Munda, and Darja Fiser. 2026. Thematic Landscapes of the Past: Analysing Slovene Historical Periodicals With Topic Modeling. In Proceedings of the First Workshop on Creating Interoperable Corpora of Historical Newspapers, pages 92–99, Palma de Mallorca, Spain. Association for Computational Linguistics.
Cite (Informal):
Thematic Landscapes of the Past: Analysing Slovene Historical Periodicals With Topic Modeling (Dobranić et al., PressMint 2026)
Copy Citation: