Elamur Research⌕
☰
home / papers / 2026 / ForumOccitania: A Corpus of User-Generated Content for Multiple Occitan Varieties

ForumOccitania: A Corpus of User-Generated Content for Multiple Occitan Varieties

Summary

Occitan is a low-resource dialect continuum covering southern France and some valleys in Spain and Italy. Existing NLP datasets for it lean heavily towards Lengadocian and usually offer only coarse dialect tags. To address this, the authors built ForumOccitania, a corpus collected from an Occitan discussion forum that was active from 2007 to 2012 and covered many topics. Each post inherits metadata from its author's voluntary profile: declared dialect, age group, normalised location and Occitan proficiency. The corpus also includes automatic language predictions, since other languages and code-switching appear. Personal data is removed through an anonymisation pipeline. Statistical and manual analyses show that most content is Occitan written in the classical spelling, mostly by young users. Posts show strong traits of Lemosin, Lengadocian, Gascon and Provençau. An unsupervised topic-modelling method from Kuparinen and Scherrer (2024) recovers these varieties, aligns well with user-declared labels and hints at finer regional patterns. The resource broadens coverage of under-served varieties and supports variation-aware NLP and corpus linguistics.

Key points

  • Source and availability. The data comes from occitania.forumactif.com (posts from 2007–2012). The code is at github.com/DEFI-COLaF/forum-occitania. The paper was published at LREC 2026 by a team from Inria Paris.
  • Metadata. The profile dialect options were Gascon, Lemosin, Lengadocian, Provençau, Vivarés and Auvernhat. Ages are grouped into buckets to protect privacy. Locations are normalised to country, administrative region (before and after the 2016 mergers), cultural region, département and locality. Proficiency is mapped to a 6-level scale.
  • Questionable labels. The Auvernhat and Vivarés labels appear to be wrong, possibly because of a website bug. For example, an "Auvernhat" post shows Gascon markers such as the enunciative que, deishar and ua.
  • Dialect identification. The authors used NMF topic models on mixed word and character n-gram features. They scored them against declared dialects on the test split using homogeneity, completeness and V-score. With eight topics, Gascon splits into two topics, one weighted towards the Pyrenean article eth.
  • Gap filled. The corpus is rich in Lemosin and Provençau, which are scarce elsewhere. Earlier resources include Tolosa Treebank, the Lo Congrès corpora and OcWikiDisc, which was the only prior Occitan user-generated content dataset.

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

Fields: Linguistics & languages

Cite

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.63317/3h3c8smpj9p2

@inproceedings{nedey2026forumoccitania,
  title = {{ForumOccitania: A Corpus of User-Generated Content for Multiple Occitan Varieties}},
  author = {Nédey, Oriane and Janès, Juliette and Bawden, Rachel and Clérice, Thibault and Sagot, Benoît},
  booktitle = {Proceedings of the Fifteenth Language Resources and Evaluation Conference},
  year = {2026},
  doi = {10.63317/3h3c8smpj9p2},
  url = {https://doi.org/10.63317/3h3c8smpj9p2}
}

References

45 references, 4 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Allison Adams, Eric Aili, Daniel Aioanei, Rebecca Jonsson, Lina Mickelsson, Dagmar Mikmekova, Fred Roberts, Javier Fernandez Valencia, and Roger Wechsler. 2019. AnonyMate: A Toolkit for Anonymizing Unstructured Chat Data. In Proceedings of the Workshop on NLP and Pseudonymisation, pages 1–7, Turku,… find
  2. Sarah C. Brennan. 2024. 21 La situation sociolinguistique de l’occitan du début du XXe siècle à nos jours. In Louise Esher and Jean Sibille, editors, Manuel de linguistique occitane, pages 593–621. De Gruyter. find
  3. David Britain. 2017. Dialect Contact and New Dialect Formation. In The Handbook of Dialectology, pages 143–158. John Wiley & Sons, Ltd. find
  4. J. K. Chambers and Peter Trudgill. 1998. Dialectology. Cambridge University Press. Google-Books-ID: 9bYV43UhKssC. find
  5. Marta R. Costa-jussà, James Cross, Onur Çelebi, Maha Elbayad, Kenneth Heafield, Kevin Heffernan, Elahe Kalbassi, Janice Lam, Daniel Licht, Jean Maillard, Anna Sun, Skyler Wang, Guillaume Wenzek, Al Youngblood, Bapi Akula, Loic Barrault, Gabriel Mejia Gonzalez, Prangthip Hansanti, John Hoffman, Semar… find
  6. Marko Dragojevic, Jessica Gasiorek, and Howard Giles. 2015. Communication Accommodation Theory. In The International Encyclopedia of Interpersonal Communication, pages 1–21. John Wiley & Sons, Ltd. find
  7. Fahim Faisal, Orevaoghene Ahia, Aarohi Srivastava, Kabir Ahuja, David Chiang, Yulia Tsvetkov, and Antonios Anastasopoulos. 2024. DIALECT-BENCH: An NLP Benchmark for Dialects, Varieties, and Closely-Related Languages. In Proceedings of the 62nd Annual Meeting of the Association for Computational Ling… find
  8. Zachary Hopton and Noëmi Aepli. 2024. Modeling Orthographic Variation in Occitan’s Dialects. In Proceedings of the Eleventh Workshop on NLP for Similar Languages, Varieties, and Dialects (VarDial 2024), pages 78–88, Mexico City, Mexico. Association for Computational Linguistics. find
  9. Dirk Hovy and Christoph Purschke. 2018. Capturing Regional Variation with Distributed Place Representations and Geographic Retrofitting. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pages 4383–4394, Brussels, Belgium. Association for Computational Lingui… find
  10. Juliette Janès, Rachel Bawden, Thibault Clérice, Rasul Dent, Lucence Ing, Oriane Nédey, and Benoît Sagot. 2025. Encoding language diversity in TEI: a description of regional and non-standard languages in multilingual diachronic corpora. In TEI Conference 2025. find
  11. Aditya Joshi, Raj Dabre, Diptesh Kanojia, Zhuang Li, Haolan Zhan, Gholamreza Haffari, and Doris Dippold. 2025. Natural Language Processing for Dialects of a Language: A Survey. ACM Comput. Surv., 57(6):149:1–149:37. find
  12. Armand Joulin, Edouard Grave, Piotr Bojanowski, and Tomas Mikolov. 2017. Bag of tricks for efficient text classification. In Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 2, Short Papers, pages 427–431, Valencia, Spain. Associatio… find
  13. Philipp Koehn, Hieu Hoang, Alexandra Birch, Chris Callison-Burch, Marcello Federico, Nicola Bertoldi, Brooke Cowan, Wade Shen, Christine Moran, Richard Zens, Chris Dyer, Ondřej Bojar, Alexandra Constantin, and Evan Herbst. 2007. Moses: Open source toolkit for statistical machine translation. In Proc… find
  14. Sneha Kudugunta, Isaac Caswell, Biao Zhang, Xavier Garcia, Derrick Xin, Aditya Kusupati, Romi Stella, Ankur Bapna, and Orhan Firat. 2023. MADLAD-400: a multilingual and document-level large audited dataset. In Proceedings of the 37th International Conference on Neural Information Processing Systems,… find our card
  15. Olli Kuparinen and Yves Scherrer. 2023. Dialect Representation Learning with Neural Dialect-to-Standard Normalization. In Tenth Workshop on NLP for Similar Languages, Varieties and Dialects (VarDial 2023), Dubrovnik, Croatia. Association for Computational Linguistics. find
  16. Olli Kuparinen and Yves Scherrer. 2024. Corpus-based dialectometry with topic models. Journal of Linguistic Geography, 12(1):1–12. find
  17. Xavier Lamuela. 2024. 20 Codification et élaboration linguistiques. In Louise Esher and Jean Sibille, editors, Manuel de linguistique occitane, pages 563–589. De Gruyter. find
  18. Hervé Lieutard. 2024. Provençal et niçard : des configurations exemplaires des tendances intégratives et séparatrices en domaine occitan. In Les noms des variantes de langue minoritaire Études de cas en France et en Russie, presses universitaires de bordeaux edition, Diglossi@_2, pages 99–117 [en li… find
  19. Jean Léo Léonard, Guylaine Brun-Trigaud, and Flore Picard. 2024. 17 Atlas linguistiques et perspectives dialectométriques. In Louise Esher and Jean Sibille, editors, Manuel de linguistique occitane, pages 473–520. De Gruyter. find
  20. Aleksandra Miletic and Yves Scherrer. 2022. OcWikiDisc: a Corpus of Wikipedia Talk Pages in Occitan. In Proceedings of the Ninth Workshop on NLP for Similar Languages, Varieties and Dialects, pages 70–79, Gyeongju, Republic of Korea. find
  21. Aleksandra Miletić. 2023. Outiller l’occitan : nouvelles ressources et lemmatisation. In Actes de CORIA-TALN 2023. Actes de la 30e Conférence sur le Traitement Automatique des Langues Naturelles (TALN), volume 1 : travaux de recherche originaux – articles longs, pages 217–231, Paris, France. ATALA. find
  22. Aleksandra Miletić, Myriam Bras, Marianne Vergez-Couret, Louise Esher, Clamença Poujade, and Jean Sibille. 2020. A Four-Dialect Treebank for Occitan: Building Process and Parsing Experiments. In Proceedings of the 7th Workshop on NLP for Similar Languages, Varieties and Dialects, pages 140–149, Barc… find
  23. Iñigo Morcillo, Igor Leturia, Ander Corral, Xabier Sarasola, Michaël Barret, Aure Séguier, and Benaset Dazéas. 2024. Automatic Speech Recognition for Gascon and Languedocian Variants of Occitan. In Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resource… find
  24. NLLB Team, Marta R. Costa-jussà, James Cross, Onur Çelebi, Maha Elbayad, Kenneth Heafield, Kevin Heffernan, Elahe Kalbassi, Janice Lam, Daniel Licht, Jean Maillard, Anna Sun, Skyler Wang, Guillaume Wenzek, Al Youngblood, Bapi Akula, Loic Barrault, Gabriel Mejia Gonzalez, Prangthip Hansanti, John Hof… find
  25. Sebastian Nordhoff and Harald Hammarström. 2011. Glottolog/Langdoc: Defining Dialects, Languages, and Language Families as Collections of Resources. In Proceedings of ISWC 2011, Bonn, Germany. find
  26. Oriane Nédey, Rachel Bawden, Thibault Clérice, and Benoît Sagot. 2026. OcWikiDialects: A Wikipedia Dataset with Rich Metadata for Occitan Dialect Identification. In Proceedings of the 13th Workshop on NLP for Similar Languages, Varieties and Dialects, Rabat, Morocco. find our card
  27. Pentti Paatero and Unto Tapper. 1994. Positive matrix factorization: A non-negative factor model with optimal utilization of error estimates of data values. Environmetrics, 5(2):111–126. find
  28. Guilherme Penedo, Hynek Kydlíček, Vinko Sabolčec, Bettina Messmer, Negar Foroutan, Amir Hossein Kargaran, Colin Raffel, Martin Jaggi, Leandro Von Werra, and Thomas Wolf. 2025. FineWeb2: One Pipeline to Scale Them All – Adapting Pre-Training Data Processing to Every Language. ArXiv:2506.20920 [cs]. find
  29. Ildikó Pilán, Pierre Lison, Lilja Øvrelid, Anthi Papadopoulou, David Sánchez, and Montserrat Batet. 2022. The Text Anonymization Benchmark (TAB): A Dedicated Corpus and Evaluation Framework for Text Anonymization. Computational Linguistics, 48(4):1053–1101. find
  30. Clamenca Poujade, Myriam Bras, and Assaf Urieli. 2024. CorpusArièja: Building an Annotated Corpus with Variation in Occitan. In Proceedings of the 3rd Annual Meeting of the Special Interest Group on Under-resourced Languages @ LREC-COLING 2024, pages 66–71, Torino, Italia. ELRA and ICCL. find
  31. Achim Rabus and Yves Scherrer. 2025. Dialects, Topic Models, and Border Effects: The Rusyn Case. In Proceedings of the 10th Workshop on Slavic Natural Language Processing (Slavic NLP 2025), pages 38–43, Vienna, Austria. Association for Computational Linguistics. find
  32. Arij Riabi, Menel Mahamdi, Virginie Mouilleron, and Djamé Seddah. 2024. Cloaked Classifiers: Pseudonymization Strategies on Sensitive Classification Tasks. In Proceedings of the Fifth Workshop on Privacy in Natural Language Processing, pages 123–136, Bangkok, Thailand. Association for Computational… find
  33. Yves Scherrer, Rob van der Goot, and Petter Mæhlum. 2025. Findings of the VarDial Evaluation Campaign 2025: The NorSID Shared Task on Norwegian Slot, Intent and Dialect Identification. In Proceedings of the 12th Workshop on NLP for Similar Languages, Varieties and Dialects, pages 1–8, Abu Dhabi, UAE… find
  34. Holger Schwenk, Vishrav Chaudhary, Shuo Sun, Hongyu Gong, and Francisco Guzmán. 2019. WikiMatrix: Mining 135M Parallel Sentences in 1620 Language Pairs from Wikipedia. ArXiv:1907.05791 [cs]. find our card
  35. Jean Sibille. 2024. 16 Les dialectes occitans. In Louise Esher and Jean Sibille, editors, Manuel de linguistique occitane, pages 423–471. De Gruyter. find
  36. Hugo Sousa, Rúben Almeida, Purificação Silvano, Inês Cantante, Ricardo Campos, and Alípio Jorge. 2025. Enhancing Portuguese variety identification with cross-domain approaches. In Proceedings of the Thirty-Ninth AAAI Conference on Artificial Intelligence and Thirty-Seventh Conference on Innovative A… find
  37. Marianne Vergez-Couret, Myriam Bras, Aleksandra Miletić, and Clamença Poujade. 2024. Loflòc: A Morphological Lexicon for Occitan using Universal Dependencies. In Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024),… find
  38. Urchade Zaratiana, Nadi Tomeh, Pierre Holat, and Thierry Charnois. 2024. GLiNER: Generalist Model for Named Entity Recognition using Bidirectional Transformer. In Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technol… find our card
  39. Özlem Çetinoğlu and Antje Schweitzer. 2022. Anonymising the SAGT Speech Corpus and Treebank. In Proceedings of the Thirteenth Language Resources and Evaluation Conference, pages 5557–5564, Marseille, France. European Language Resources Association. 9. find
  40. Language Resource References Omar Kamali. 2025. Wikipedia Monthly. Omneity Labs. HuggingFace datasets, 1.0. find
  41. Séguier, Aure and Lo Congrès. 2023a. Occitan Corpus from Lo Congrès news. Lo Congrès permanent de la lenga occitana. Lo Congrès, distributed via Zenodo: DOI 10.5281/zen-odo.8411197, 1.0. — [doi:10.5281/zen-odo.8411197](https://doi.org/10.5281/zen-odo.8411197) find
  42. Séguier, Aure and Lo Congrès. 2023b. SoftwaresOccitanTranslations corpus. Lo Congrès permanent de la lenga occitana. Lo Congrès, distributed via Zenodo: DOI 10.5281/zen-odo.8411351, 1.0. — [doi:10.5281/zen-odo.8411351](https://doi.org/10.5281/zen-odo.8411351) find
  43. Séguier, Aure and Lo Congrès. 2024. Lo Congrès websites Corpus. Lo Congrès permanent de la lenga occitana. Lo Congrès, distributed via Zenodo: DOI 10.5281/zenodo.12192029, 1.0. — [doi:10.5281/zenodo.12192029](https://doi.org/10.5281/zenodo.12192029) find
  44. Vergez-Couret, Marianne and Bras, Myriam and Miletić, Aleksandra and Poujade, Clamença. 2024. Loflòc: A Morphological Lexicon for Occitan using Universal Dependencies. 1.0. find
  45. Zeman, Daniel. 2025. Universal Dependencies 2.17. Universal Dependencies Consortium, 2.17. find

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/nedey-2026-forumoccitania-corpus-user-generated-content/