EtymoLink: A Structured English Etymology Dataset

What it is about

The authors turn the prose etymologies of the Online Etymology Dictionary (Etymonline) into structured relations between word roots. They hand-annotate a subset of entries as ground truth and fine-tune FLAN-T5-base to extract the relations for the rest. The resulting dataset has over 103,000 relationships over 63,603 English terms, with 15,931 connected components. On the root-relationship evaluation the system reaches 94.4% accuracy. Language labels are harder: detection is 0.990, but label accuracy is 0.909. Output in the right format covers 98.2% of entries once the language-label constraint is dropped. The authors also report that more training data mainly helps the label-based metrics.

Ideas

• I1. en/src/morph.rs and en/src/dict.rs: load the paper's relations (root, affix, source language) as a table keyed by lemma, and use it as a check on the lemma and affix analysis. Measure how many lemmas in the EWT test set get a relation, and how often the morph analysis agrees with the relation.
• I2. Deterministic extractor: write a Rust pattern matcher over the Etymonline-style phrases ("from X, Y, Z") in a raw dump kept in $MOVA_DATA, and compare its root accuracy with the paper's 94.4% on a hand-checked sample of 200 entries. This turns the neural step into a readable rule set.
• I3. Connected-component test: count the components and the most connected affixes (the paper's top five are all affixes) and check that en/src/morph.rs treats those affixes as affixes, not as roots.

For Mova

The touch points are en/src/morph.rs, en/src/dict.rs and a rule matcher in tools/. The chapters are 16. Derivation I: prefixes and 17. Derivation II: suffixes, with 09. Early Modern and archaic forms for the old source languages. The dataset is CC BY 4.0, but it is derived from Etymonline, so check that source's terms before loading it into the build. Measure first the agreement of the morph analysis with the relations on the EWT lemmas.

Cite

The paper itself, not this card. BibTeX from the library's record of the work. DOI 10.18653/v1/2024.lchange-1.12

@inproceedings{gao2024etymolink,
  title = {{EtymoLink: A Structured English Etymology Dataset}},
  author = {Gao, Yuan and Sun, Weiwei},
  booktitle = {LChange 2024 (ACL workshop, Bangkok)},
  year = {2024},
  doi = {10.18653/v1/2024.lchange-1.12},
  url = {https://doi.org/10.18653/v1/2024.lchange-1.12}
}

References

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

1. Robert K. Barnhart and Sol Steinmetz. 1988. The Barnhart Dictionary of Etymology. H.W. Wilson Company. find
2. Khuyagbaatar Batsuren, Gabor Bella, and Fausto Giunchiglia. 2019. CogNet: A Large-Scale Cognate Database. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 3136–3145, Florence, Italy. Association for Computational Linguistics. find
3. Albert C. Baugh and Thomas Cable. 1993. A History of the English Language. Taylor & Francis. find
4. Shane Bergsma and Grzegorz Kondrak. 2007. Alignment-Based Discriminative String Similarity. In Proceedings of the 45th Annual Meeting of the Association of Computational Linguistics, pages 656–663, Prague, Czech Republic. Association for Computational Linguistics. find
5. Marcel Bollmann. 2018. Normalization of historical texts with neural network models. find
6. Hyung Won Chung, Le Hou, Shayne Longpre, Barret Zoph, Yi Tay, William Fedus, Yunxuan Li, Xuezhi Wang, Mostafa Dehghani, Siddhartha Brahma, Albert Webson, Shixiang Shane Gu, Zhuyun Dai, Mirac Suzgun, Xinyun Chen, Aakanksha Chowdhery, Alex Castro-Ros, Marie Pellat, Kevin Robinson, Dasha Valter, Sharan Narang, Gaurav Mishra, Adams Yu, Vincent Zhao, Yanping Huang, Andrew Dai, Hongkun Yu, Slav Petrov, Ed H. Chi, Jeff Dean, Jacob Devlin, Adam Roberts, Denny Zhou, Quoc V. Le, and Jason Wei. 2022. Scaling Instruction-Finetuned Language Models. Preprint, arXiv:2210.11416. ↗
7. Alina Maria Ciobanu and Liviu P. Dinu. 2014. Automatic Detection of Cognates Using Orthographic Alignment. In Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), pages 99–105, Baltimore, Maryland. Association for Computational Linguistics. find
8. Gerard de Melo. 2014. Etymological Wordnet: Tracing The History of Words. In Proceedings of the Ninth International Conference on Language Resources and Evaluation (LREC’14), pages 1148–1154, Reykjavik, Iceland. European Language Resources Association (ELRA). find our card
9. Ernest Klein. 1971. A Comprehensive Etymological Dictionary Of The English Language By Ernest Klein. find
10. Piotr G asiorowski. 2006. The etymology of Old English *docga. 111:275–284. find
11. Luís Gomes and José Gabriel Pereira Lopes. 2011. Measuring Spelling Similarity for Cognate Identification. In Progress in Artificial Intelligence, pages 624–633, Berlin, Heidelberg. Springer. find
12. Koustava Goswami, Priya Rani, Theodorus Fransen, and John McCrae. 2023. Weakly-supervised Deep Cognate Detection Framework for Low-Resourced Languages Using Morphological Knowledge of Closely-Related Languages. In Findings of the Association for Computational Linguistics: EMNLP 2023, pages 531–541, Singapore. Association for Computational Linguistics. find
13. Diptesh Kanojia, Kevin Patel, Malhar Kulkarni, Pushpak Bhattacharyya, and Gholemreza Haffari. 2019. Utilizing Wordnets for Cognate Detection among Indian Languages. In Proceedings of the 10th Global Wordnet Conference, pages 404–412, Wroclaw, Poland. Global Wordnet Association. find
14. Chin-Yew Lin. 2004. ROUGE: A Package for Automatic Evaluation of Summaries. In Text Summarization Branches Out, pages 74–81, Barcelona, Spain. Association for Computational Linguistics. find
15. Johann-Mattis List, Simon J. Greenhill, Cormac Anderson, Thomas Mayer, Tiago Tresoldi, and Robert Forkel. 2018. CLICS2: An improved database of cross-linguistic colexifications assembling lexical data with the help of cross-linguistic data formats. Linguistic Typology, 22(2):277–306. find
16. Christian M. Meyer and Iryna Gurevych. 2012. Wiktionary: A new rival for expert-built lexicons? Exploring the possibilities of collaborative lexicography. In Electronic Lexicography, pages 259–292. Oxford University Press. find
17. George A. Miller. 1995. WordNet: A lexical database for English. Communications of the ACM, 38(11):39– 41. ↗ our card
18. Ruslan Mitkov, Viktor Pekar, Dimitar Blagoev, and Andrea Mulloni. 2007. Methods for extracting and classifying pairs of cognates and false friends. Machine Translation, 21(1):29–53. find
19. Roberto Navigli and Simone Paolo Ponzetto. 2012. BabelNet: The automatic construction, evaluation and application of a wide-coverage multilingual semantic network. Artificial Intelligence, 193:217–250. find
20. Arika Okrent and Sean O’Neill. 2021. Blame the Printing Press. In Arika Okrent and Sean O’Neill, editors, Highly Irregular: Why Tough, Through, and Dough Don’t Rhyme—And Other Oddities of the English Language, page 0. Oxford University Press. find
21. Kishore Papineni, Salim Roukos, Todd Ward, and Wei-Jing Zhu. 2002. Bleu: A Method for Automatic Evaluation of Machine Translation. In Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics, pages 311–318, Philadelphia, Pennsylvania, USA. Association for Computational Linguistics. find our card
22. Maja Popović. 2015. chrF: Character n-gram F-score for automatic MT evaluation. In Proceedings of the Tenth Workshop on Statistical Machine Translation, pages 392–395, Lisbon, Portugal. Association for Computational Linguistics. find our card
23. Michel Simard, George F. Foster, and Pierre Isabelle. 1992. Using cognates to align sentences in bilingual corpora. In Proceedings of the Fourth Conference on Theoretical and Methodological Issues in Machine Translation of Natural Languages, Montréal, Canada. find
24. J. A. Simpson and E. S. C. Weiner. 1989. The Oxford English Dictionary. Clarendon Press. find
25. Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Ł ukasz Kaiser, and Illia Polosukhin. 2017. Attention is All you Need. In Advances in Neural Information Processing Systems, volume 30. Curran Associates, Inc. find

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/lex-2024-gao-etymolink-english/