Elamur Research⌕
☰
home / papers / 2026 / The Indonesian Religiolect Corpus: Data Curation for Muslim, Protestant, and Catholic Language Varieties

The Indonesian Religiolect Corpus: Data Curation for Muslim, Protestant, and Catholic Language Varieties

Paper · 2026 · Dan Sachs

Summary

Indonesian is well-resourced overall, but varieties shaped by religious identity ("religiolects") lack dedicated data and benchmarks. This leaves them effectively low-resource inside a high-resource language. The author builds the Indo-Religiolect Corpus, about 3 million sentences scraped from more than 100 institutional websites of Muslim, Catholic, and Protestant communities. The pipeline uses trusted-node scraping, balanced undersampling, and validation by classification.

Fine-tuning IndoBERT to predict the community behind a sentence reaches 88.26% overall. Separability is asymmetric: Islamic Indonesian stands out clearly, while Catholic and Protestant text overlaps heavily in vocabulary but still carries detectable markers.

The author argues that models trained on generic, majority-normative web text may blur these minority varieties, a "representation collapse." This could plausibly create fairness risks in moderation, hiring, translation, and sentiment tools, although downstream tasks are not tested. The work offers a reusable template for documenting sub-national varieties, moving beyond single national-language benchmarks.

Key points

  • Corpus size and sources: ~3M sentences in total. The Catholic portion has 797,131 sentences from ~30 sites (e.g., Mirifica, KAS), and the Protestant portion has 760,587 sentences from ~44 sites (e.g., PGI). The text is mostly from 2020–2025.
  • IndoBERT classification: 88.26% overall, with per-class results of 91.73% for Islam, 86.41% for Catholic, and 86.64% for Protestant.
  • Linguistic markers discussed:
    • Arabic-derived vocabulary and Kitab Malay–influenced syntax in Muslim text
    • Arabic suffix variants such as -at vs. -ah
    • Differing Allah/Tuhan semantics between Muslims and Christians
    • English loanwords in Protestant writing
  • Limitations: The corpus covers only formal, institutional web register; informal and spoken usage, smaller faiths, and finer denominations are left for future work. The author also flags dual-use risks such as religious profiling.
  • Availability: Data and code are at github.com/dansachs/indo-religiolects.

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

🎓Universitas Gadjah MadaYogyakarta · Indonesia4 works · 56 citations

 Fields, topics and concepts

📚 Shelves 4The 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 works

Cite

Cite as Dan Sachs. 2026. The Indonesian Religiolect Corpus: Data Curation for Muslim, Protestant, and Catholic Language Varieties. Proceedings of the Second Workshop on Language Models for Low-Resource Languages (LoResLM 2026).

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2026.loreslm-1.36

@inproceedings{sachs2026indonesian,
  title = {{The Indonesian Religiolect Corpus: Data Curation for Muslim, Protestant, and Catholic Language Varieties}},
  author = {Dan Sachs},
  booktitle = {Proceedings of the Second Workshop on Language Models for Low-Resource Languages (LoResLM 2026)},
  year = {2026},
  doi = {10.18653/v1/2026.loreslm-1.36},
  url = {https://doi.org/10.18653/v1/2026.loreslm-1.36}
}

References

22 references, 7 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Alessandro Bausani. 1975. Is classical malay a “muslim language”? Boletin de la Asociacion Española de Orientalistas. find
  2. Damián Blasi, Antonios Anastasopoulos, and Graham Neubig. 2022. Systematic inequalities in language technology performance across the world’s languages. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 5486–5505. Association fo…
  3. Su Lin Blodgett, Solon Barocas, Hal Daumé III, and Hanna Wallach. 2020. Language (technology) is power: A critical survey of “bias” in NLP. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 5454–5476. Association for Computational Linguistics.
  4. Samuel Cahyawijaya, Holy Lovenia, Alham Fikri Aji, Genta Indra Winata, Bryan Wilie, Rahmad Mahendra, Christian Wibisono, Ade Romadhony, Karissa Vincentio, Fajri Koto, Jennifer Santoso, David Moeljadi, Cahya Wirawan, Frederikus Hudi, Ivan Halim Parmonangan, Ika Alfina, Muhammad Satrio Wicaksono, Ilha… find
  5. 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, and 20 o… find
  6. J. N. Pava, C. Meinhardt, H. B. U. Zaman, and T. Friedman. 2024. Mind the (language) gap: Mapping the challenges of llm development in low-resource language contexts. Stanford HAI Policy Brief. find
  7. Peter G. Riddell. 2002. Literal translation, sacred scripture and kitab malay. Studia Islamika, 9(1). find
  8. Jayanta Sadhu, Maneesha Rani Saha, and Rifat Shahriyar. 2024. Social bias in large language models for bangla: An empirical study on gender and religious bias. Preprint, arXiv:2407.03536. find
  9. Maarten Sap, Dallas Card, Saadia Gabriel, Yejin Choi, and Noah A. Smith. 2019. The risk of racial bias in hate speech detection. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 1668–1678. Association for Computational Linguistics.
  10. Emanuel Gerrit Singgih. 2003. Doing Theology in Indonesia: Sketches for an Indonesian Contextual Theology. ATESEA. find
  11. Stuart Campbell. 1996. The distribution of -at and -ah endings in malay loanwords from arabic. Bijdragen tot de Taal-, Land- en Volkenkunde, 152(1):23–44. find
  12. Lucas Dixon, John Li, Jeffrey Sorensen, Nithum Thain, and Lucy Vasserman. 2018. Measuring and mitigating unintended bias in text classification. In Proceedings of the 2018 AAAI/ACM Conference on AI, Ethics, and Society, pages 67–73. find
  13. Jonathan Dunn. 2019. Modeling global syntactic variation in english using dialect classification. In Proceedings of the Sixth Workshop on NLP for Similar Languages, Varieties and Dialects, pages 42–53. Association for Computational Linguistics.
  14. Charles A. Ferguson. 1959. Diglossia. Word, 15(2):325– 340. find
  15. Kevin W. Fogg. 2015. The standardisation of the indonesian language and its consequences for islamic communities. Journal of Southeast Asian Studies, 46(1):86–110. find
  16. Benjamin Hary and Martin J. Wein. 2013. Religiolin-guistics: On jewish-, christian- and muslim-defined languages. International Journal of the Sociology of Language, 2013(220):85–108. find
  17. Fajri Koto, Afshin Rahimi, Jey Han Lau, and Timothy Baldwin. 2020. Indolem and indobert: A benchmark dataset and pre-trained language model for indonesian nlp. In Proceedings of the 28th International Conference on Computational Linguistics, pages 757– 770. International Committee on Computational L… find
  18. James N. Sneddon. 2003. The Indonesian Language: Its History and Role in Modern Society. University of New South Wales Press. find
  19. Nikolaos Van Dam. 2010. Arabic loanwords in indonesian revisited. Bijdragen tot de Taal-, Landen Volkenkunde, 166(2/3):218–243. find
  20. Genta Indra Winata, Alham Fikri Aji, Samuel Cahyawijaya, Rahmad Mahendra, Fajri Koto, Ade Romadhony, Kemal Kurniawan, David Moeljadi, Radityo Eko Prasojo, Pascale Fung, Timothy Baldwin, Jey Han Lau, Rico Sennrich, and Sebastian Ruder. 2023. Nusax: Multilingual parallel sentiment dataset for 10 indon…
  21. Marcos Zampieri, Shervin Malmasi, Nikola Ljubešić, Preslav Nakov, Ahmed Ali, Jörg Tiedemann, Yves Scherrer, and Noëmi Aepli. 2017. Findings of the VarDial evaluation campaign 2017. In Proceedings of the Fourth Workshop on NLP for Similar Languages, Varieties and Dialects (VarDial). Association for…
  22. Marcos Zampieri, Shervin Malmasi, Preslav Nakov, Ahmed Ali, Suwon Shon, James Glass, Yves Scherrer, Tanja Samardžić, Nikola Ljubešić, Jörg Tiedemann, Chris van der Lee, Stefan Grondelaers, Nelleke Oostdijk, Dirk Speelman, Antal van den Bosch, Ritesh Kumar, Bornini Lahiri, and Mayank Jain. 2018. La… — Language Identification and Morphosyntactic Tagging: The Second VarDial Evaluation Campaign

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/sachs-2026-indonesian-religiolect-corpus-data-curation/