---
title: "Proffiliadur — Gutiérrez-Rolón 2026"
description: "Readability tools exist mainly for English and other well-resourced languages."
authors: ["Nicolás Gutiérrez-Rolón", "Jonathan Davies", "Tomos Williams", "Dawn Knight", "Fernando Alva-Manchego"]
year: 2026
venue: "Proceedings of the Fifteenth Language Resources and Evaluation Conference"
source: "https://aclanthology.org/2026.lrec-1.88/"
doi: "10.63317/5c6yawn79s5h"
licence: "CC BY-NC 4.0 (ELRA, LREC proceedings)"
licence_url: "https://creativecommons.org/licenses/by-nc/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/papers/rolon-2026-proffiliadur-welsh-language-text-profiling/"
---
[Elamur Research](https://md.research.elamur.ai/) · [this page in HTML](https://research.elamur.ai/free/papers/rolon-2026-proffiliadur-welsh-language-text-profiling/)

[home](https://md.research.elamur.ai/) / [papers](https://md.research.elamur.ai/free/papers/) / [2026](https://md.research.elamur.ai/free/papers/2026/page/4/) / Proffiliadur: Welsh Language Text Profiling Toolkit

# Proffiliadur: Welsh Language Text Profiling Toolkit

**Authors:** [Nicolás Gutiérrez-Rolón](https://md.research.elamur.ai/free/authors/nicolas-gutierrez-rolon/), [Jonathan Davies](https://md.research.elamur.ai/free/authors/jonathan-davies/), Tomos Williams, [Dawn Knight](https://md.research.elamur.ai/free/authors/dawn-knight/), [Fernando Alva-Manchego](https://md.research.elamur.ai/free/authors/fernando-alva-manchego/) · **Year:** 2026 · **Venue:** Proceedings of the Fifteenth Language Resources and Evaluation Conference · **Licence:** [CC BY-NC 4.0 (ELRA, LREC proceedings)](https://creativecommons.org/licenses/by-nc/4.0/)

**Links:** [Publisher's page](https://aclanthology.org/2026.lrec-1.88/) · [PDF](https://aclanthology.org/2026.lrec-1.88.pdf) · [DOI 10.63317/5c6yawn79s5h](https://doi.org/10.63317/5c6yawn79s5h) · [OpenAlex](https://openalex.org/W7160099512)

**Citations:** 0 citations (openalex-snapshot · 2026-10-14) · [2](#cited-by) in our library, of the 140673 papers whose bibliographies we read

## Summary
Readability tools exist mainly for English and other well-resourced languages. Welsh, a minority and low-resource language, has had no dedicated way to measure how complex or accessible a text is. Its morphology, such as initial consonant mutation and short-form verbs, also makes length-based formulae unreliable. The authors built Proffiliadur, an open-source Python toolkit that extracts 141 interpretable indices covering surface, lexical, morphological and syntactic properties. It uses a Welsh-specific tokenisation step, the CyTag2 tagger, a UDPipe Universal Dependencies parser and Welsh vocabulary lists. They tested it in two ways. First, they contrasted texts written under the Cymraeg Clîr ("Clear Welsh") guidance with ordinary Welsh texts. Second, they profiled learner-oriented texts at different CEFR levels. They also compared feature-based, embedding-based and neural classifiers for separating "clear" from "not clear" texts. The toolkit's features alone scored F1 = 0.94, close to a fine-tuned transformer at F1 = 0.97, while being cheaper to run and easier to interpret. The work gives Welsh public bodies, educators and health communicators a reproducible way to check plain-language compliance at scale.

## Key points
- **Feature set:** 141 indices built from counts, means, unique incidences and ratios.
  - Surface statistics include 11 indices, among them an adapted Gulpease readability score.
  - Further groups cover lexical profiling, PoS statistics and dependency-tree metrics (depth, breadth, head distance, branching direction).
  - Welsh-specific mutation counts cover soft, nasal and aspirate mutation and h-prothesis.
- **External tools:** CyTag2, a rule-based tagger from the CorCenCC project, handles tokenisation and morphology. UDPipe's Welsh UD model handles parsing.
- **Lexical resources:**
  - Yr Amliadur frequency bands K1–K6+, each band holding 1,000 words.
  - WJEC/CBAC CEFR word lists: 675 words for A1, 652 for A2 and 756 for B1.
  - Two stopword lists: 437 Welsh-only words and 486 words including English.
- **Classification result:** linguistic features reached F1 = 0.94 versus 0.97 for a fine-tuned transformer. The study also prompted Claude with Clear Welsh criteria as an LLM comparison.
- **Release and findings:** the code is available at github.com/cardiffnlp/proffiliadur. The analyses offer empirical backing for Cymraeg Clîr recommendations involving impersonal verb forms, explicit subjects, shorter clauses and periphrastic constructions.

*Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.*

Fields: [Linguistics & languages](https://md.research.elamur.ai/free/shelves/linguistics/)

## Cited by

Cited by

2 papers of our library cite it (of the 140673 whose bibliographies we read); 1 of them have a page here, 1 not yet.

- [CEFR-Cymraeg: A Dataset and Baseline Models for Language Proficiency Assessment in Welsh](https://md.research.elamur.ai/free/papers/waqar-2026-cefr-cymraeg-dataset-baseline-models/)

## Cite

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

```bibtex
@inproceedings{gutierrezrolon2026proffiliadur,
  title = {{Proffiliadur: Welsh Language Text Profiling Toolkit}},
  author = {Gutiérrez-Rolón, Nicolás and Davies, Jonathan and Williams, Tomos and Knight, Dawn and Alva-Manchego, Fernando},
  booktitle = {Proceedings of the Fifteenth Language Resources and Evaluation Conference},
  year = {2026},
  doi = {10.63317/5c6yawn79s5h},
  url = {https://doi.org/10.63317/5c6yawn79s5h}
}
```

- [← previous Probing the Dynamics of Syntactic Ability Acquisition Throughout LLM Pretraining](https://md.research.elamur.ai/free/papers/matsuda-2026-probing-dynamics-syntactic-ability-acquisition/)
- [next → Proficiency-Controlled Text Simplification in European Portuguese: A Preliminary Study using Prompting Approaches](https://md.research.elamur.ai/free/papers/ribeiro-2026-proficiency-controlled-text-simplification-european/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/papers/2026/page/4/)

## References

31 references, 1 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Kepa Bengoetxea and Itziar Gonzalez-Dios. 2021. Multiaztertest: a multilingual analyzer on multiple levels of language for readability assessment. [find](https://scholar.google.com/scholar?q=Kepa+Bengoetxea+and+Itziar+Gonzalez%2DDios%2E+2021%2E+Multiaztertest%3A+a+multilingual+analyzer+on+multiple+levels+of+language+for+readability+assessment%2E)
2. Daniel Cunliffe, Andreas Vlachidis, Daniel Williams, and Douglas Tudhope. 2022. Natural language processing for under-resourced languages: Developing a welsh natural language toolkit. Computer Speech & Language, 72:101311. [find](https://scholar.google.com/scholar?q=Daniel+Cunliffe%2C+Andreas+Vlachidis%2C+Daniel+Williams%2C+and+Douglas+Tudhope%2E+2022%2E+Natural+language+processing+for+under%2Dresourced+languages%3A+Developing+a+welsh+natural+language+toolkit%2E+Computer+Speech+)
3. Mrinmoy Dalal et al. 2023. Language agnostic readability assessments. In Proceedings of the 2023 IEEE Conference on Software Engineering and Computer Engineering (CSCE), pages 252– 257. IEEE. [find](https://scholar.google.com/scholar?q=Mrinmoy+Dalal+et+al%2E+2023%2E+Language+agnostic+readability+assessments%2E+In+Proceedings+of+the+2023+IEEE+Conference+on+Software+Engineering+and+Computer+Engineering+%28CSCE%29%2C+pages+252%E2%80%93+257%2E+IEEE%2E)
4. Ignatius Ezeani, Mahmoud El-Haj, Jonathan Morris, and Dawn Knight. 2022. Introducing the Welsh text summarisation dataset and baseline systems. In Proceedings of the Thirteenth Language Resources and Evaluation Conference, pages 5097–5106, Marseille, France. European Language Resources Association. [find](https://scholar.google.com/scholar?q=Ignatius+Ezeani%2C+Mahmoud+El%2DHaj%2C+Jonathan+Morris%2C+and+Dawn+Knight%2E+2022%2E+Introducing+the+Welsh+text+summarisation+dataset+and+baseline+systems%2E+In+Proceedings+of+the+Thirteenth+Language+Resources+and+)
5. D. B. Friedman, L. Hoffman-Goetz, and J. F. Arocha. 2006. Health literacy and the world wide web: Comparing the readability of leading incident cancers on the internet. Medical Informatics and the Internet in Medicine, 31(1):67–87. [find](https://scholar.google.com/scholar?q=D%2E+B%2E+Friedman%2C+L%2E+Hoffman%2DGoetz%2C+and+J%2E+F%2E+Arocha%2E+2006%2E+Health+literacy+and+the+world+wide+web%3A+Comparing+the+readability+of+leading+incident+cancers+on+the+internet%2E+Medical+Informatics+and+the+Int)
6. Itziar Gonzalez Dios. 2016. Readability Assessment and Automatic Text Simplification: The Analysis of Basque Complex Structures. Ph.D. thesis, University of the Basque Country, Basque Language and Communication Department. CC BY-NC-SA 4.0. [find](https://scholar.google.com/scholar?q=Itziar+Gonzalez+Dios%2E+2016%2E+Readability+Assessment+and+Automatic+Text+Simplification%3A+The+Analysis+of+Basque+Complex+Structures%2E+Ph%2ED%2E+thesis%2C+University+of+the+Basque+Country%2C+Basque+Language+and+Com)
7. Joseph Marvin Imperial. 2021. Bert embeddings for automatic readability assessment. ArXiv preprint. [find](https://scholar.google.com/scholar?q=Joseph+Marvin+Imperial%2E+2021%2E+Bert+embeddings+for+automatic+readability+assessment%2E+ArXiv+preprint%2E)
8. Joseph Marvin Imperial, Abdullah Barayan, Regina Stodden, Rodrigo Wilkens, Ricardo Muñoz Sánchez, Lingyun Gao, Melissa Torgbi, Dawn Knight, Gail Forey, Reka R. Jablonkai, Ekaterina Kochmar, Robert Joshua Reynolds, Eugénio Ribeiro, Horacio Saggion, Elena Volodina, Sowmya Vajjala, Thomas François, Fer… [find](https://scholar.google.com/scholar?q=Joseph+Marvin+Imperial%2C+Abdullah+Barayan%2C+Regina+Stodden%2C+Rodrigo+Wilkens%2C+Ricardo+Mu%C3%B1oz+S%C3%A1nchez%2C+Lingyun+Gao%2C+Melissa+Torgbi%2C+Dawn+Knight%2C+Gail+Forey%2C+Reka+R%2E+Jablonkai%2C+Ekaterina+Kochmar%2C+Robert+Jos)
9. Dawn Knight, Steven Morris, Tess Fitzpatrick, Paul Rayson, Irena Spasić, Enlli Môn Thomas, Alex Lovell, Jonathan Morris, Jeremy Evas, Mark Stonelake, Laura Arman, Joshua Davies, Ignatius Ezeani, Steven Neale, Jennifer Needs, Scott Piao, Mair Rees, Gareth Watkins, Lowri Williams, Vignesh Muralidaran,… [find](https://scholar.google.com/scholar?q=Dawn+Knight%2C+Steven+Morris%2C+Tess+Fitzpatrick%2C+Paul+Rayson%2C+Irena+Spasi%C4%87%2C+Enlli+M%C3%B4n+Thomas%2C+Alex+Lovell%2C+Jonathan+Morris%2C+Jeremy+Evas%2C+Mark+Stonelake%2C+Laura+Arman%2C+Joshua+Davies%2C+Ignatius+Ezeani%2C+Steve)
10. Dawn Knight, Steven Morris, Tony Fitzpatrick, Paul Rayson, Irena Spasić, and Enlli M. Thomas. 2020b. The national corpus of contemporary welsh: Project report \| y corpws cenedlaethol cymraeg cyfoes: Adroddiad y prosiect. ArXiv preprint arXiv:2010.05542. — [arXiv:2010.05542](https://arxiv.org/abs/2010.05542) [find](https://scholar.google.com/scholar?q=Dawn+Knight%2C+Steven+Morris%2C+Tony+Fitzpatrick%2C+Paul+Rayson%2C+Irena+Spasi%C4%87%2C+and+Enlli+M%2E+Thomas%2E+2020b%2E+The+national+corpus+of+contemporary+welsh%3A+Project+report+%5C%7C+y+corpws+cenedlaethol+cymraeg+cyfoes%3A+)
11. Dawn Knight, Stuart Morris, Bethan Tovey-Walsh, Tom Fitzpatrick, and Laura Anthony. 2020c. Yr amliadur: Frequency lists for contemporary welsh. Technical report, Cardiff University. [find](https://scholar.google.com/scholar?q=Dawn+Knight%2C+Stuart+Morris%2C+Bethan+Tovey%2DWalsh%2C+Tom+Fitzpatrick%2C+and+Laura+Anthony%2E+2020c%2E+Yr+amliadur%3A+Frequency+lists+for+contemporary+welsh%2E+Technical+report%2C+Cardiff+University%2E)
12. Ivan Kraus. 2025. Predicting readability of czech legal writing using linguistic features. Bachelor’s thesis defended on June 9, 2025. [find](https://scholar.google.com/scholar?q=Ivan+Kraus%2E+2025%2E+Predicting+readability+of+czech+legal+writing+using+linguistic+features%2E+Bachelor%E2%80%99s+thesis+defended+on+June+9%2C+2025%2E)
13. Sidney Evaldo Leal, Magali Sanches Duran, Carolina Evaristo Scarton, Nathan Siegle Hartmann, and Sandra Maria Aluísio. 2024. Nilc-metrix: assessing the complexity of written and spoken language in brazilian portuguese. Language Resources and Evaluation, 58(1):73–110. [find](https://scholar.google.com/scholar?q=Sidney+Evaldo+Leal%2C+Magali+Sanches+Duran%2C+Carolina+Evaristo+Scarton%2C+Nathan+Siegle+Hartmann%2C+and+Sandra+Maria+Alu%C3%ADsio%2E+2024%2E+Nilc%2Dmetrix%3A+assessing+the+complexity+of+written+and+spoken+language+in+bra)
14. Tim Loughran and Bill McDonald. 2015. Measuring readability in financial disclosures. Journal of Financial Economics, 118(2):258–278. [find](https://scholar.google.com/scholar?q=Tim+Loughran+and+Bill+McDonald%2E+2015%2E+Measuring+readability+in+financial+disclosures%2E+Journal+of+Financial+Economics%2C+118%282%29%3A258%E2%80%93278%2E)
15. P. Lucisano and Maria E. Piemontese. 1988. Gulpease: una formula per la predizione della difficoltà dei testi in lingua italiana. Scuola e Città, 39(3):110–124. In Italian. [find](https://scholar.google.com/scholar?q=P%2E+Lucisano+and+Maria+E%2E+Piemontese%2E+1988%2E+Gulpease%3A+una+formula+per+la+predizione+della+difficolt%C3%A0+dei+testi+in+lingua+italiana%2E+Scuola+e+Citt%C3%A0%2C+39%283%29%3A110%E2%80%93124%2E+In+Italian%2E)
16. Emilio Matricciani. 2023. Readability indices do not say it all on a text readability. Analytics, 2(2):296– 314. [find](https://scholar.google.com/scholar?q=Emilio+Matricciani%2E+2023%2E+Readability+indices+do+not+say+it+all+on+a+text+readability%2E+Analytics%2C+2%282%29%3A296%E2%80%93+314%2E)
17. Danielle S. McNamara, Arthur C. Graesser, Philip M. McCarthy, and Zhiqiang Cai. 2014. Automated Evaluation of Text and Discourse with Coh-Metrix. Cambridge University Press. [find](https://scholar.google.com/scholar?q=Danielle+S%2E+McNamara%2C+Arthur+C%2E+Graesser%2C+Philip+M%2E+McCarthy%2C+and+Zhiqiang+Cai%2E+2014%2E+Automated+Evaluation+of+Text+and+Discourse+with+Coh%2DMetrix%2E+Cambridge+University+Press%2E)
18. Steven Neale, Kevin Donnelly, Gareth Watkins, and Dawn Knight. 2018. Leveraging lexical resources and constraint grammar for rule-based part-of-speech tagging in welsh. In Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC 2018), pages 3946–3954, Miyazaki… [find](https://scholar.google.com/scholar?q=Steven+Neale%2C+Kevin+Donnelly%2C+Gareth+Watkins%2C+and+Dawn+Knight%2E+2018%2E+Leveraging+lexical+resources+and+constraint+grammar+for+rule%2Dbased+part%2Dof%2Dspeech+tagging+in+welsh%2E+In+Proceedings+of+the+Eleventh+)
19. OECD. 2011. National survey of adult skills 2011. [https://www.oecd.org/skills/nationalsurveyofadultskills.htm](https://www.oecd.org/skills/nationalsurveyofadultskills.htm). Accessed 13 October 2025. [↗](https://www.oecd.org/skills/nationalsurveyofadultskills.htm)
20. Lloyd Lois Antonie Reyes, Michael Antonio Ibañez, Ranz Sapinit, Mohammed Hussien, and Joseph Marvin Imperial. 2022. A baseline readability model for cebuano. In Proceedings of the 17th Workshop on Innovative Use of NLP for Building Educational Applications (BEA 2022). Association for Computational L… [find](https://scholar.google.com/scholar?q=Lloyd+Lois+Antonie+Reyes%2C+Michael+Antonio+Iba%C3%B1ez%2C+Ranz+Sapinit%2C+Mohammed+Hussien%2C+and+Joseph+Marvin+Imperial%2E+2022%2E+A+baseline+readability+model+for+cebuano%2E+In+Proceedings+of+the+17th+Workshop+on+Inn)
21. Tanya Serry, Tonya Stebbins, Andrew Martchenko, Natalie Araujo, and Brigid McCarthy. 2022. Improving access to covid-19 information by ensuring the readability of government websites. Health Promotion Journal of Australia, 34(1):68– 72. [find](https://scholar.google.com/scholar?q=Tanya+Serry%2C+Tonya+Stebbins%2C+Andrew+Martchenko%2C+Natalie+Araujo%2C+and+Brigid+McCarthy%2E+2022%2E+Improving+access+to+covid%2D19+information+by+ensuring+the+readability+of+government+websites%2E+Health+Promotion)
22. Milan Straka and Jana Straková. 2017. Udpipe: Trainable pipeline for processing conll-u files performing tokenization, morphosyntactic tagging, lemmatization and dependency parsing. In Proceedings of the CoNLL 2017 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies, pages 88–9… [find](https://scholar.google.com/scholar?q=Milan+Straka+and+Jana+Strakov%C3%A1%2E+2017%2E+Udpipe%3A+Trainable+pipeline+for+processing+conll%2Du+files+performing+tokenization%2C+morphosyntactic+tagging%2C+lemmatization+and+dependency+parsing%2E+In+Proceedings+of+)
23. Cíntia Matsuda Toledo, Sandra Maria Aluísio, Leandro Borges dos Santos, Sonia Maria Dozzi Brucki, Eduardo Sturzeneker Trés, Maira Okada de Oliveira, and Letícia Lessa Mansur. 2018. Analysis of macrolinguistic aspects of narratives from individuals with alzheimer’s disease, mild cognitive impairment,… [find](https://scholar.google.com/scholar?q=C%C3%ADntia+Matsuda+Toledo%2C+Sandra+Maria+Alu%C3%ADsio%2C+Leandro+Borges+dos+Santos%2C+Sonia+Maria+Dozzi+Brucki%2C+Eduardo+Sturzeneker+Tr%C3%A9s%2C+Maira+Okada+de+Oliveira%2C+and+Let%C3%ADcia+Lessa+Mansur%2E+2018%2E+Analysis+of+macroli)
24. S. Uchida. 2025. Assigning cefr-j levels to english learners’ writing. Journal of Second Language Writing, S2772766125000205. [find](https://scholar.google.com/scholar?q=S%2E+Uchida%2E+2025%2E+Assigning+cefr%2Dj+levels+to+english+learners%E2%80%99+writing%2E+Journal+of+Second+Language+Writing%2C+S2772766125000205%2E)
25. Tiffany M. Walsh and Teresa A. Volsko. 2008. Readability assessment of internet-based consumer health information. Respiratory Care, 53(10):1310–1315. [find](https://scholar.google.com/scholar?q=Tiffany+M%2E+Walsh+and+Teresa+A%2E+Volsko%2E+2008%2E+Readability+assessment+of+internet%2Dbased+consumer+health+information%2E+Respiratory+Care%2C+53%2810%29%3A1310%E2%80%931315%2E)
26. Zhi Wang, Yifan Zhang, Zhi Li, and Zhiwei Liu. 2019. Readability assessment of textbooks in low resource languages. volume 61, pages 23109– 23121. [find](https://scholar.google.com/scholar?q=Zhi+Wang%2C+Yifan+Zhang%2C+Zhi+Li%2C+and+Zhiwei+Liu%2E+2019%2E+Readability+assessment+of+textbooks+in+low+resource+languages%2E+volume+61%2C+pages+23109%E2%80%93+23121%2E)
27. Eeshan Waqar, Jonathan Davies, Dawn Knight, and Fernando Alva-Manchego. 2026. Cefr-cymraeg: A dataset and baseline models for language proficiency assessment in welsh. In Proceedings of the Fifteenth Language Resources and Evaluation Conference, Palma, Mallorca. European Language Resources Associati… [find](https://scholar.google.com/scholar?q=Eeshan+Waqar%2C+Jonathan+Davies%2C+Dawn+Knight%2C+and+Fernando+Alva%2DManchego%2E+2026%2E+Cefr%2Dcymraeg%3A+A+dataset+and+baseline+models+for+language+proficiency+assessment+in+welsh%2E+In+Proceedings+of+the+Fifteenth+) [our card](https://md.research.elamur.ai/free/papers/waqar-2026-cefr-cymraeg-dataset-baseline-models/)
28. Welsh Government. 2017. Cymraeg 2050: A million welsh speakers. [https://www.gov.wales/cymraeg-2050-welsh-language-strategy](https://www.gov.wales/cymraeg-2050-welsh-language-strategy). Accessed 13 October 2025. [↗](https://www.gov.wales/cymraeg-2050-welsh-language-strategy)
29. Welsh Government. 2018. Welsh language technology action plan. [https://www.gov.wales/welsh-language-technology-action-plan](https://www.gov.wales/welsh-language-technology-action-plan). Accessed 13 October 2025. [↗](https://www.gov.wales/welsh-language-technology-action-plan)
30. C. Williams. 1999. Cymraeg Clir. Cyngor Gwynedd. [find](https://scholar.google.com/scholar?q=C%2E+Williams%2E+1999%2E+Cymraeg+Clir%2E+Cyngor+Gwynedd%2E)
31. Full list of Features Table 2: Full List of Toolkit Indices, with Subclasses and explanations Feature Subclass General Text Statistics and Readability Indices Sentence Count Basic Token Count Basic Number Count Basic Word Count Basic Average Sentence Length Basic Unique Words Basic TTR (Type Token R… [find](https://scholar.google.com/scholar?q=Full+list+of+Features+Table+2%3A+Full+List+of+Toolkit+Indices%2C+with+Subclasses+and+explanations+Feature+Subclass+General+Text+Statistics+and+Readability+Indices+Sentence+Count+Basic+Token+Count+Basic+Nu)