Mova0.5 pre-release⌕
☰
home / docs / papers / Semitic languages / At the Crossroad of Cuneiform and NLP: Challenges for Fine-grained Part-of-speech Tagging

At the Crossroad of Cuneiform and NLP: Challenges for Fine-grained Part-of-speech Tagging

What it is about

The authors prepare a corpus of 121 Old Babylonian Akkadian letters from Sippar, annotated by three specialists with PoS tags, citation forms and morphemes in the ORACC style. They fine-tune FLAIR sequence taggers that use different pre-trained embeddings: multilingual, Arabic, Spanish and Japanese. The study asks three questions: whether the embeddings from a related language help, how to cut the text into sentences when no syntax is annotated, and whether the input should be transliteration or Unicode cuneiform. Verb-final sentence splitting gives 94.8% PoS accuracy in five-fold cross-validation, against 92.5% for splitting at physical lines. Feeding Unicode cuneiform instead of transliteration costs between 19.5 and 30.7 points, depending on the embedding. The paper ends by arguing that every text format carries some human reading decision.

Ideas

For Mova

I1 is a rule for the clause boundaries in Syntax: the clause and the sentence, measured on treebanks/hbo/ptnk. I2 is a required test in 1. Script and pointing. I3 is a design note for treebanks/akk, where no Akkadian treebank yet exists.

Topics: Akkadian and cuneiform NLP · Part-of-speech tagging

Cite

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

@inproceedings{smidt2024at,
  title = {{At the Crossroad of Cuneiform and NLP: Challenges for Fine-grained Part-of-speech Tagging}},
  author = {Smidt, Gustav Ryberg and Lefever, Els and Graef, Katrien de},
  booktitle = {Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024)},
  year = {2024},
  url = {https://aclanthology.org/2024.lrec-main.154/}
}

References

27 references, 2 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Ž. Agić and I. Vulić. 2019. JW300: A wide-coverage parallel corpus for low-resource languages. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 3204–3210, Florence, Italy. Association for Computational Linguistics. find
  2. R. Bansal, H. Choudhary, R. Punia, N. Schenk, J.L. Dahl, and É. Pagé-Perron. 2021. How low is too low? a computational perspective on extremely low-resource languages. In Proceedings of the Joint Conference of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing: Student Research Workshop, pages 44–59. find
  3. C. Chiarcos, I. Khait, É. Pagé-Perron, N. Schenk, Jayanth, C. Fäth, J. Steuers, W. Mcgrath, and J. Wang. 2018. Annotating a low-resource language with llod technology: Sumerian morphology and syntax. Information, 9(11). find
  4. W. de Vries, M. Wieling, and M. Nissim. 2022. Make the best of cross-lingual transfer: Evidence from POS tagging with over 100 languages. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 7676–7685, Dublin, Ireland. Association for Computational Linguistics. find
  5. G. Deutscher. 2000. Syntactic Change in Akkadian: The Evolution of Sentential Complementation. Oxford University Press. find
  6. R. Frankena. 1966. Briefe aus dem British Museum (LIH und CT 2-33). Number 2 in Altbabylonische Briefe in Umschrift und Übersetzung. E. J. Brill. find
  7. S. Gordin, G. Gutherz, A. Elazary, A. Romach, E. Jiménez, J. Berant, and Y. Cohen. 2020. Reading akkadian cuneiform using natural language processing. PLoS ONE, 15(10). find
  8. S. Gordin and A. Romach. 2022. Optical character recognition for complex scripts: A case-study in cuneiform. In Digital Humanities 2022: Responding to Asian Diversity, pages 212–215. find
  9. G. Gutherz, S. Gordin, L. Sáenz, O. Levy, and J. Berant. 2023. Translating akkadian to english with neural machine translation. PNAS Nexus, 2(5):1–10. find
  10. J. Huehnergard. 2011. A Grammar of Akkadian. Eisenbrauns. find
  11. L.W. King. 1900. The Letters and Inscriptions of Hammurabi, King of Babylon, About B.C. 2200. Vol. II. Number 3 in Luzac’s Semitic Text and Translation Series. Luzac And Co. find
  12. G. Klein, Y. Kim, Y. Deng, J. Senellart, and A. Rush. 2017. Opennmt: Open-source toolkit for neural machine translation. In Proceedings of ACL 2017, System Demonstrations, pages 67–72. find
  13. F.R. Kraus and K.R. Veenhof, editors. 1964–2005. Albabylonische Briefe in Umschrift und Übersetzung. 14 vols. Brill. find
  14. M. Luukko, A. Sahala, S. Hardwick, and K. Lindén. 2020. Akkadian treebank for early Neo-Assyrian royal inscriptions. In Proceedings of the 19th International Workshop on Treebanks and Linguistic Theories, pages 124–134, Düsseldorf, Germany. Association for Computational Linguistics. find our card
  15. M. Ong and S. Gordin. 2024. Linguistic annotation of cuneiform texts using treebanks and deep learning. Digital Scholarship in the Humanities, pages 1–12. find
  16. E. Reiner. 1966. A Linguistic Analysis of Akkadian. Number 21 in Janua Linguarum, Series Practica. Mouton & Co. find
  17. E. Reiner. 1973. How we read cuneiform texts. Journal of Cuneiform Studies, 25(1):3–58. find
  18. A. Sahala. 2021. Contributions to Computational Assyriology. Ph.D. thesis, University of Helsinki. find
  19. A. Sahala and K. Lindén. 2023. Babylemmatizer 2.0 – a neural pipeline for pos-tagging and lemmatizing cuneiform languages. In Proceedings of the Ancient Language Processing Workshop associated with RANLP-2023, pages 203–212. find
  20. A. Sahala, M. Silfverberg, A. Arppe, and K. Lindén. 2020. BabyFST - towards a finite-state based computational model of ancient babylonian. In Proceedings of the 12th Language Resources and Evaluation Conference, pages 3886–3894, Marseille, France. European Language Resources Association. find our card
  21. P. Singh, A. Maladry, and E. Lefever. 2023. Too many cooks spoil the model: Are bilingual models for Slovene better than a large multilingual model? In Proceedings of the 9th Workshop on Slavic Natural Language Processing 2023 (SlavicNLP 2023), pages 32–39, Dubrovnik, Croatia. Association for Computational Linguistics. find
  22. T. Sommerschield, Y. Assael, J. Pavlopoulos, V. Stefanak, A. Senior, C. Dyer, J. Bodel, J. Prag, I. Androutsopoulos, and N. de Freitas. 2023. Machine learning for ancient languages: A survey. Computational Linguistics, pages 1–45. find
  23. M.P. Streck. 2010. Großes fach altorientalistik: Der umfang des keilschriftlichen textkorpus. Mitteilungen der Deutschen Orient-Gesellschaft zu Berlin, 142:35–58. find
  24. M.P. Streck. 2022. Old Babylonian Grammar, Vol. 1. Number 168.1 in Handbook of Oriental Studies, Section One: Ancient Near East. Brill. find
  25. M. Sukhareva, F. Fuscagni, J. Daxenberger, S. Görke, D. Prechel, and I. Gurevych. 2017. Distantly supervised pos tagging of low-resource languages under extreme data sparsity: The case of hittite. In Joint SIGHUM Workshop on Computational Linguistics for Cultural Heritage, Social Sciences, Humanities and Literature, Proceedings, pages 95–104, Vancouver, Canada. find
  26. C. Swaelens, I. De Vos, and E. Lefever. 2023. Medieval social media: Manual and automatic annotation of byzantine Greek marginal writing. In Proceedings of the 17th Linguistic Annotation Workshop (LAW-XVII), pages 1–9, Toronto, Canada. Association for Computational Linguistics. find
  27. P. Zemánek. 2007. A treebank of ugaritic. annotating fragmentary attested languages. In Proceedings of the Sixth International Workshop on Treebanks and Linguistic Theories. 10. Appendices 10.1. Results Related to Section 5.1, 5.2 and 5.3 Embeddings Multilingual Arabic Spanish Japanese Accuracy 71,0% 76,2% 74,1% 72,6% find

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/coling-2024-smidt-crossroad-cuneiform-nlp-challenges-fine/