Elamur Research⌕
☰
home / papers / 2026 / Historical Newspapers in the General Regionally Annotated Corpus of Ukrainian (GRAC): Current State and PressMint Integration Prospects

Historical Newspapers in the General Regionally Annotated Corpus of Ukrainian (GRAC): Current State and PressMint Integration Prospects

Paper · 2026 · Maria Shvedova and Arsenii Lukashevskyi

Summary

The paper describes the pre-1950 newspaper component of GRAC, which serves as the de facto national corpus of Ukrainian, and assesses what it would take to bring it into PressMint, CLARIN's infrastructure for interoperable historical press corpora. Ukrainian newspapers from this era are difficult to process. Their territories were split among several states, regional literary variants and competing spelling systems coexisted, and the texts often mix in other languages.

The authors outline where the collection comes from, how unevenly it is spread across years and macroregions, and how its metadata records the controlling political entity and regional attribution. They show that tools built for modern Ukrainian, namely TagText and UDPipe2, struggle with the Galician Zhelekhivka orthography, mostly because of its divergent spelling and word segmentation.

Comparing GRAC's vertical files for NoSketch Engine with PressMint's TEI-based format, they find that conversion is technically feasible. Success, however, depends on normalizing metadata, developing taxonomies and adapting annotation layers. Integration would let researchers study Ukrainian press alongside material from neighboring European contexts.

Key points

  • The historical subset covers 117 newspaper titles published before 1950, about 23.6 million tokens of human-checked OCR text. WWII is the richest period, with roughly 4.4M tokens.
  • Sources include Orest Drul's Zbruch archive (Galician press), the LIBRARIA scans and the Archive of Old Newspapers (processed by students), and a WWII set compiled by Anna Bordovska. The WWII set spans occupation, Soviet and OUN-UPA underground papers.
  • The doc.mediaAdmin attribute assigns each periodical one of 12 political-entity codes, for example AHI, RUI, POL, SOV, RUK and OUN. Pre-WWII papers carry article-level metadata, while most later papers were digitised as whole issues.
  • Manual checks of UDPipe2 POS tags showed a clear gap. On a 572-token sample from Soviet Ukrainian Bil'shovyk Poltavshchyny (1924), accuracy was 96.7%. For Dilo, an 1889 Lviv paper written in Zhelekhivka, a sample of 504 tokens yielded only 92.7%.
  • Planned steps include:
    • treating Zhelekhivka texts as a separate subcorpus with a dedicated UDPipe2 model;
    • converting named entities into PressMint's four classes (PERSON, LOCATION, ORGANIZATION, MISC).

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Institutions

🎓Friedrich Schiller University JenaJena · Germany3 works · 187 citations

Also on the byline: National Technical University "Kharkiv Polytechnic Institute".

 Fields, topics and concepts

📚 Shelves 5The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksArtificial Intelligence · 16,194 worksArts and Humanities · 16,039 worksLanguage and Linguistics · 15,891 worksLiterature and Literary Theory · 519 works

Cite

Cite as Maria Shvedova and Arsenii Lukashevskyi. 2026. Historical Newspapers in the General Regionally Annotated Corpus of Ukrainian (GRAC): Current State and PressMint Integration Prospects. Proceedings of the First Workshop on Creating Interoperable Corpora of Historical Newspapers.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.63317/4zhmbuez4k25

@inproceedings{shvedova2026historical,
  title = {{Historical Newspapers in the General Regionally Annotated Corpus of Ukrainian (GRAC): Current State and PressMint Integration Prospects}},
  author = {Maria Shvedova and Arsenii Lukashevskyi},
  booktitle = {Proceedings of the First Workshop on Creating Interoperable Corpora of Historical Newspapers},
  year = {2026},
  doi = {10.63317/4zhmbuez4k25},
  url = {https://doi.org/10.63317/4zhmbuez4k25}
}

References

13 references, 0 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. 2010–2016. Archive of old newspapers. Digital archive of historical Ukrainian newspapers. find
  2. Maria Shvedova. 2021. Lexical variation in the language of the Ukrainian press of the 1920s– 1940s and the development of a new lexical norm: A corpus-based research. Movoznavstvo, (1):16– 35. In Ukrainian. find
  3. Arkhivni Informatsijni Systemy. 2017–. LIBRARIA: Archive of Ukrainian periodicals. find
  4. Anna Bordovska. 2024. Orthographic and lexical variation in the journalism of the Second World War period: Based on GRAC data. Movni i kontseptual’ni kartyny svitu, 1(75):36–59. In Ukrainian. find
  5. Maria Shvedova and Ruprecht von Waldenfels. 2021. Regional annotation within GRAC, a large reference corpus of Ukrainian: Issues and challenges. In Proceedings of the 5th International Conference on Computational Linguistics and Intelligent Systems (COLINS), pages 32–45, Kharkiv, Ukraine. find
  6. Yurij Chemerys, Olesia Nakhlik, Andriy Rysin, and Maria Shvedova. 2023. Normalization of a historic Western Ukrainian orthographic system Zhelekhivka in the Ukrainian language reference corpus (GRAC). In Proceedings of the IEEE 18th International Conference on Computer Sciences and Information Techn… find
  7. Vasyl Starko and Andriy Rysin. 2022. VESUM: A large morphological dictionary of Ukrainian as a dynamic tool. In Proceedings of the 6th International Conference on Computational Linguistics and Intelligent Systems (COLINS), pages 71–80, Gliwice, Poland. find
  8. CLARIN ERIC. 2025. Pressmint: Interoperable corpora of historical newspapers. Accessed: 2026-02-28. find
  9. Orest Drul. 2014–. Western Ukrainian historical press collections. Zbruč Digital Archive. Three chronological collections: 125 Years Ago (1897– 1902), 100 Years Ago (1922–1927), 75 Years Ago (1947–1952). 11. find
  10. Language Resource References Zynovija Franko. 1995. Variation or territorial distinction of the Ukrainian literary language. Ukrans’ka istorična ta dialektna leksyka, (2):169–173. In Ukrainian. find
  11. Maria Shvedova, Ruprecht von Waldenfels, Sergey Yarygin, Andriy Rysin, Vasyl Starko, Tymofij Nikolajenko, Arsenii Lukashevskyi et al. 2017–. GRAC: General Regionally Annotated Corpus of Ukrainian. find
  12. Anastasiia Hleba. 2025. Regional variation of the spatial Ukrainian prepositions in the 1920s–1930s: a corpus-based study. In Synsé-mantické slovní druhy ve slovanských jazycích, pages 147–166. Institute of Slavonic Studies of the Czech Academy of Sciences, Prague. find
  13. Milan Straka, Jana Straková and Jan Hajič. 2016– . UDPipe Web Service (LINDAT/CLARIAH-CZ): Trainable Pipeline for Tokenization, Tagging, Lemmatization and Parsing. LINDAT/CLARIAH-CZ, Institute of Formal and Applied Linguistics, Charles University. find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/shvedova-2026-historical-newspapers-general-regionally-annotated/