Cognate Detection for Historical Language Reconstruction of Proto-Sabean Languages: the Case of Ge'ez, Tigrinya, and Amharic

What it is about

The paper works on the Ethio-Semitic branch, with three languages: Ge'ez, Tigrinya and Amharic. It digitizes a three-way dictionary of 14,100 words per language, converts the words to IPA with Epitran, and runs three cognate detectors from LingPy: the Turchin consonant-class rule, the Sound Class Algorithm (SCA) and LexStat. Linguists first mark 74 cognate sets on the 100-word Swadesh list; SCA finds 81% of them and LexStat 69%. Pairs that both methods accept give 1,847 automatic cognate sets, and linguists check a sample of 196, finding errors in 2%. A few-shot GPT-4o reconstructs the proto-forms of the 74 expert sets with 85% accuracy, and its outputs are used as training data for mT5 and AfriTeVa with LoRA. On the expert-reconstructed set, mT5 reaches 57.7% and AfriTeVa 12.2%.

Ideas

• I1. Deterministic cognate gate: the Turchin rule (two words are candidates when their first two consonant classes match) is a table lookup. Apply it to Hebrew, Aramaic and Arabic forms of the same gloss, and measure its precision and recall against a small expert-marked set in the style of the 74 sets.
• I2. Audit protocol: keep only the pairs that two independent methods accept, then check a sample by a linguist and report the error rate. Use the same protocol before any auto-built root or cognate list enters the Mova Semitic lexicon.
• I3. Sound-class alignment for roots: map the consonants of Hebrew and Aramaic lemmas to sound classes and align them with SCA-style scoring, to find forms that share a root. This is a candidate route to root data for the macula dict lemmas, which have none yet.

For Mova

I1 and I2 go into The lexicon of Biblical Aramaic and The lexicon, with the test set in tools/macula. I3 is a plan for root data in macula dict and for the cross-variety comparison in The Aramaic family and the Jewish varieties.

Cite

The paper itself, not this card. BibTeX from the library's record of the work.

@inproceedings{temesgen2025cognate,
  title = {{Cognate Detection for Historical Language Reconstruction of Proto-Sabean Languages: the Case of Ge'ez, Tigrinya, and Amharic}},
  author = {Temesgen, Elleni Sisay and Nigatu, Hellina Hailu and Andargie, Fitsum Assamnew},
  booktitle = {Proceedings of the 31st International Conference on Computational Linguistics},
  year = {2025},
  url = {https://aclanthology.org/2025.coling-main.496/}
}

References

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

1. Solomon Teferra Abate, Martha Yifiru Tachbelie, Michael Melese, Hafte Abera, Tewodros Abebe, Wondwossen Mulugeta, Yaregal Assabie, Million Meshesha, Solomon Afnafu, and Binyam Ephrem Seyoum. 2020. Large vocabulary read speech corpora for four Ethiopian languages: Amharic, Tigrigna, Oromo and Wolaytta. In Proceedings of the Twelfth Language Resources and Evaluation Conference, pages 4167–4171, Marseille, France. European Language Resources Association. find
2. Henok Ademtew and Mikiyas Birbo. 2024. AGE: Amharic, Ge’ez and English parallel dataset. In Proceedings of the Seventh Workshop on Technologies for Machine Translation of Low-Resource Languages (LoResMT 2024), pages 139–145, Bangkok, Thailand. Association for Computational Linguistics. find
3. Zra Dawit Adhana. 1995. Lsanat Sem. Kidst Slase Menfesawi Kolej, addis ababa. find
4. VSDS Akavarapu and Arnab Bhattacharya. 2024. Automated cognate detection as a supervised link prediction task with cognate transformer. arXiv preprint arXiv:2402.02926. ↗
5. Raimo Anttila. 1989. Historical and comparative linguistics. find
6. Alexandre Bouchard-Côté, Thomas L Griffiths, and Dan Klein. 2009. Improved reconstruction of protolanguage word forms. In Proceedings of human language technologies: The 2009 annual conference of the north american chapter of the association for computational linguistics, pages 65–73. find
7. Alina Maria Ciobanu, Liviu P Dinu, and Laurentiu Zoicas. 2020. Automatic reconstruction of missing romanian cognates and unattested latin words. In Proceedings of the Twelfth Language Resources and Evaluation Conference, pages 3226–3231. find
8. A. Dolgopolsky. 1964. Gipoteza drevnejšego rodstva jazykovych semej severnoj evrazii s verojatnostej točky zrenija [a probabilistic hypothesis concering the oldest relationships among the language families of northern eurasia]. find
9. EKI Archive. n.d. Rom2-ti document. https://arhiiv.eki.ee/wgrs/v2_2/rom2_ti.pdf. Accessed: 2024-07-30. ↗
10. Robert Hetzron, Alan S Kaye, and Ghil’ad Zuckermann. 2018. Semitic languages. In The World’s Major Languages, pages 568–576. Routledge. find
11. Edward J. Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu, Yuanzhi Li, Shean Wang, Lu Wang, and Weizhu Chen. 2021. LoRA: Low-Rank Adaptation of Large Language Models. arXiv. find
12. John Huehnergard et al. 2013. The Semitic Languages. Routledge. find
13. Odunayo Jude Ogundepo, Akintunde Oladipo, Mofetoluwa Adeyemi, Kelechi Ogueji, and Jimmy Lin. 2022. AfriTeVA: Extending ?small data? pretraining approaches to sequence-to-sequence models. In Proceedings of the Third Workshop on Deep Learning for Low-Resource Natural Language Processing, pages 126–135, Hybrid. Association for Computational Linguistics. find
14. Young Min Kim, Kalvin Chang, Chenxuan Cui, and David Mortensen. 2023. Transformed protoform reconstruction. arXiv preprint arXiv:2307.01896. ↗
15. Vladimir I Levenshtein et al. 1966. Binary codes capable of correcting deletions, insertions, and reversals. In Soviet physics doklady, volume 10, pages 707–710. Soviet Union. find
16. Johann-Mattis List. 2010. Sca: Phonetic alignment based on sound classes. In European Summer School in Logic, Language and Information, pages 32–51. Springer. find
17. Johann-Mattis List. 2012. LexStat: Automatic detection of cognates in multilingual wordlists. In Proceedings of the EACL 2012 Joint Workshop of LINGVIS & UN-CLH, pages 117–125, Avignon, France. Association for Computational Linguistics. find
18. Johann-Mattis List, Simon J Greenhill, and Russell D Gray. 2017. The potential of automatic word comparison for historical linguistics. PloS one, 12(1):e0170046. find
19. Johann-Mattis List, Ekaterina Vylomova, Robert Forkel, Nathan Hill, and Ryan Cotterell. 2022. The sigtyp 2022 shared task on the prediction of cognate reflexes. In Proceedings of the 4th Workshop on Research in Computational Linguistic Typology and Multilingual NLP, pages 52–62. Association for Computational Linguistics. find
20. Roddy MacSween and Andrew Caines. 2020. An expectation maximisation algorithm for automated cognate detection. In Proceedings of the 24th Conference on Computational Natural Language Learning, pages 476–485, Online. Association for Computational Linguistics. find
21. Carlo Meloni, Shauli Ravfogel, and Yoav Goldberg. 2019. Ab antiquo: Neural proto-language reconstruction. arXiv preprint arXiv:1908.02477. ↗
22. Carlo Meloni, Shauli Ravfogel, and Yoav Goldberg. 2021. Ab antiquo: Neural proto-language reconstruction. In Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 4460–4473, Online. Association for Computational Linguistics. find
23. Girma Neshir, Solomon Atnafu, and Andreas Rauber. 2020. Bert fine-tuning for amharic sentiment classification. In Workshop RESOURCEFUL Co-Located with the Eighth Swedish Language Technology Conference (SLTC), Gothenburg, Sweden, volume 25. find
24. Hellina Nigatu, Atnafu Tonja, and Jugal Kalita. 2023. The less the merrier? investigating language representation in multilingual models. In Findings of the Association for Computational Linguistics: EMNLP 2023, pages 12572–12589, Singapore. Association for Computational Linguistics. find
25. Hellina Hailu Nigatu, Atnafu Lambebo Tonja, Benjamin Rosman, Thamar Solorio, and Monojit Choudhury. 2024. The zeno’s paradox of ‘low-resource’ languages. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 17753–17774, Miami, Florida, USA. Association for Computational Linguistics. find
26. Turchin Peter, Peiros Ilia, and Gell-Mann Murray. 2010. Analyzing genetic connections between languages by matching consonant classes. , (5 (48)):117–126. find
27. Abrhalei Frezghi Tela. 2020. Sentiment analysis for low-resource language: The case of tigrinya. Master’s thesis, Itä-Suomen yliopisto. find
28. Atnafu Lambebo Tonja, Tadesse Destaw Belay, Israel Abebe Azime, Abinew Ali Ayele, Moges Ahmed Mehamed, Olga Kolesnikova, and Seid Muhie Yimam. 2023. Natural language processing in Ethiopian languages: Current state, challenges, and Parameter training_batch_size eval_batch_size epochs learning_rate lora_rank lora_dropout lora_alpha Value 32 8 100 1e-4 4 0.01 32 *Table 3: Hyperparamters for training mT5 and AfriTeVa models.* opportunities. In Proceedings of the Fourth workshop on Resources for African Indigenous Languages (RAIL 2023), pages 126–139, Dubrovnik, Croatia. Association for Computational Linguistics. find
29. Linting Xue, Noah Constant, Adam Roberts, Mihir Kale, Rami Al-Rfou, Aditya Siddhant, Aditya Barua, and Colin Raffel. 2021. mT5: A massively multilingual pre-trained text-to-text transformer. In Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 483–498, Online. Association for Computational Linguistics. find

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/coling-2025-temesgen-cognate-detection-historical-language-reconstruction/