Simple Models for Word Formation in English Slang

What it is about

The paper models three kinds of irregular word formation in English slang: blends (such as "fantabulous" from "fantastic" and "fabulous"), clippings (such as "prof" from "professor"), and reduplicatives (such as "boo-boo"). For blends it uses a character-level sequence model (CopyCat-style, with a language-model term and a length term) to choose which parts of the two source words to keep, and it reports edit distance on two datasets. On the D_knight blend data the single model scores 1.98, its ensemble 1.91, and the best prior model 1.76. For clippings it uses a phoneme-aware and a grapheme-based probabilistic model on 173 clippings, with mean edit distance 3.39 and 2.65 against a naive baseline of 4.6. For reduplicatives it uses a generative model with phonetic constraints and reports MIR on 50 held-out items; it beats the baselines by at least 8 percentage points. Most blend outputs (99.0% on D_knight) use only characters that appear in the two source words.

Ideas

• I1. Blend check in en/src/morph.rs: a candidate blend is valid only if all its characters come from the two source words, in order, and its length is at most the longer source. The paper finds 99% of blends obey this, so use the check as a rule and count the exceptions.
• I2. Reduplicative rule table: encode the patterns the paper names (duplication, and the exchange of the initial vowel or consonant) as a closed set of rules in en/src/morph.rs, each with a count of matches in the open text, so every reduplicative can say which pattern fired.
• I3. Clipping as a deterministic choice of cut point: among the cut points that keep a stressed syllable, choose the one with the highest count in Mova's open text, and report the edit distance against a fixed clipping list.
• I4. Negative control: give the blend rule two unrelated words; the rule must find no valid blend, or the character check is not doing anything.

For Mova

The chapters are 18. Compounding, conversion and the minor processes (blends, clippings, reduplicatives as minor word formation). The code is en/src/morph.rs. The measure: the rate of valid blends under the character rule, and mean edit distance on a held-out list, run twice. The paper's data and text are licensed for reading only: use the summary and the design, and build the rule tables from Mova's own open sources.

Cite

The paper itself, not this card. BibTeX from the library's record of the work. arXiv 1804.02596

@misc{kulkarni2018simple,
  title = {{Simple Models for Word Formation in English Slang}},
  author = {Kulkarni, Vivek and Wang, William Yang},
  year = {2018},
  eprint = {1804.02596},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/1804.02596}
}

References

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

1. John Algeo. 1977. Blends, a structural and systemic view. American speech 52(1/2):47–64. find
2. Joan Beal. 1991. Toy boys and lager louts: Motivation by linguistic form. In Language usage and description: Studies presented to NE Osselton on the occasion of his retirement. pages 139–148. find
3. Natalia Beliaeva. 2014. Unpacking contemporary english blends: Morphological structure, meaning, processing . find
4. Rachel Broad, Brandon Prickett, Elliott Moreton, Katya Pertsova, and Jennifer L Smith. 2016. Emergent faithfulness to proper nouns in novel english blends. In Proceedings of WCCFL. volume 33, pages 77–87. find
5. Aliya Deri and Kevin Knight. 2015. How to make a frenemy: Multitape fsts for portmanteau generation. find
6. Connie Eble. 2012. Slang and sociability: In-group language among college students. UNC Press Books. find
7. David Eppstein. 1998. Finding the k shortest paths. SIAM Journal on computing 28(2):652–673. find
8. Ingrid Fandrych. 2008. Submorphemic elements in the formation of acronyms, blends and clippings. Lexis. Journal in English Lexicology (2). find
9. Varun Gangal, Harsh Jhamtani, Graham Neubig, Eduard Hovy, and Eric Nyberg. 2017. Charmanteau: Character embedding models for portmanteau creation. arXiv preprint arXiv:1707.01176 . ↗
10. Alex Graves et al. 2012. Supervised sequence labelling with recurrent neural networks, volume 385. Springer. find
11. Stefan Th Gries. 2004a. Isnt that fantabulous? how similarity motivates intentional morphological blends in english. Language, culture, and mind pages 415–428. find
12. Stefan Th Gries. 2004b. Shouldn’t it be breakfunch? a quantitative analysis of blend structure in english. Linguistics pages 639–668. find
13. Camiel Hamans. 1997. Clippings in modern french, english, german and dutch. Trends in Linguistic Studies and Monographs 101:1733–1742. find
14. Ben Hixon, Eric Schneider, and Susan L Epstein. 2011. Phonemic similarity metrics to compare pronunciation methods. In Twelfth Annual Conference of the International Speech Communication Association. find
15. Josef Hladky. 1998. Notes on reduplicative words in english. Brno studies in English 24:33–78. find
16. Denis Jamet. 2009. A morphophonological approach to clipping in english. can the study of clipping be formalized? Lexis. Journal in English Lexicology (HS 1). find
17. Diederik Kingma and Jimmy Ba. 2014. Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980 . ↗
18. William Labov. 1972. Some principles of linguistic methodology. Language in society 1(1):97–120. find
19. Elisa Mattiello. 2005. The pervasiveness of slang in standard and non-standard english . find
20. Elisa Mattiello. 2008. An introduction to English slang: A description of its morphology, semantics and sociology, volume 2. Polimetrica sas. find
21. Elisa Mattiello. 2013. Extra-grammatical morphology in English: abbreviations, blends, reduplicatives, and related phenomena, volume 82. Walter de Gruyter. find
22. L Merlini Barbaresi. 2008. Extra-grammatical morphology: English reduplicatives. Words in action. Diachronic and synchronic approaches to English discourse. Studies in honour of Ermanno Barisone pages 228–241. find
23. Rio Rini Diah Moehkardi. 2016. Patterns and meanings of english words through word formation processes of acronyms, clipping, compound and blending found in internet-based media. Humaniora 28(3):207–221. find
24. Tadao Murata. 1990. Ab type onomatopes and reduplicatives in english and japanese. Linguistic fiesta: festschrift for Professor Hisao Kakehis sixtieth birthday pages 257–272. find
25. Gözde Özbal and Carlo Strapparava. 2012. A computational approach to the automation of creative naming. In Proceedings of the 50th Annual Meeting of the Association for Computational Linguistics: Long Papers-Volume 1. Association for Computational Linguistics, pages 703–711. find
26. Barbara Plank, Anders Søgaard, and Yoav Goldberg. 2016. Multilingual part-of-speech tagging with bidirectional long short-term memory models and auxiliary loss. arXiv preprint arXiv:1604.05529 . ↗
27. Vincent Renner, François Maniez, and Pierre Arnaud. 2013. Cross-disciplinary perspectives on lexical blending, volume 252. Walter De Gruyter. find
28. Katherine E Shaw, Andrew M White, Elliott Moreton, and Fabian Monrose. 2014. Emergent faithfulness to morphological and semantic heads in lexical blends. In Proceedings of the Annual Meetings on Phonology. volume 1. find
29. Michael R Smith, Ryan S Hintze, and Dan Ventura. 2014. Nehovah: A neologism creator nomen ipsum. find
30. Pavol Štekauer, Salvador Valera, and Lı́via Kőrtvélyessy. 2012. Word-formation in the world’s languages: A typological survey. Cambridge University Press. find
31. Nils Thun. 1963. Reduplicative words in English: A study of formations of the types tick-tick, hurly-burly, and shilly-shally. Uppsala. find
32. Friedrich Ungerer. 2007. Word-formation. In The Oxford handbook of cognitive linguistics. find
33. Peilu Wang, Yao Qian, Frank K Soong, Lei He, and Hai Zhao. 2015. Part-of-speech tagging with bidirectional long short-term memory recurrent neural network. arXiv preprint arXiv:1510.06168 . ↗
34. Kaisheng Yao and Geoffrey Zweig. 2015. Sequence-to-sequence neural net models for grapheme-to-phoneme conversion. arXiv preprint arXiv:1506.00196 . ↗
35. Jin Y Yen. 1971. Finding the k shortest loopless paths in a network. management Science 17(11):712–716. find

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/kulkarni-2018-simple-models-word-formation-english/