---
title: "AraNPCC: The Arabic Newspaper COVID-19 Corpus (2022)"
description: "Large Arabic newspaper corpora existed before the pandemic, but none covered the… 0.688 Spearman. 31 references, cited by 2 papers here."
authors: ["Abdulmohsen Al-Thubaity", "Sakhar Alkhereyf", "Alia O. Bahanshal"]
year: 2022
venue: "Proceedinsg of the 5th Workshop on Open-Source Arabic Corpora and Processing Tools with Shared Tasks on Qur'an QA and Fine-Grained Hate Speech Detection"
source: "https://aclanthology.org/2022.osact-1.4/"
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/en/papers/thubaity-2022-aranpcc-arabic-newspaper-covid-19/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/thubaity-2022-aranpcc-arabic-newspaper-covid-19/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2022](https://md.research.elamur.ai/free/en/papers/2022/page/8/) / AraNPCC: The Arabic Newspaper COVID-19 Corpus

# AraNPCC: The Arabic Newspaper COVID-19 Corpus

**Authors:** [Abdulmohsen Al-Thubaity](https://md.research.elamur.ai/free/en/authors/abdulmohsen-al-thubaity/), [Sakhar Alkhereyf](https://md.research.elamur.ai/free/en/authors/sakhar-alkhereyf/) and Alia O. Bahanshal · **Year:** 2022 · **Venue:** [Proceedinsg of the 5th Workshop on Open-Source Arabic Corpora and Processing Tools with Shared Tasks on Qur'an QA and Fine-Grained Hate Speech Detection](https://md.research.elamur.ai/free/en/venues/osact/) ([Workshop on Open-Source Arabic Corpora and Processing Tools with Shared Tasks 2022](https://md.research.elamur.ai/free/en/venues/osact/2022/)) · **Publisher:** [European Language Resources Association (ELRA)](https://md.research.elamur.ai/free/en/publishers/european-language-resources-association-elra/) · **Licence:** [CC BY-NC 4.0 (ELRA, LREC proceedings)](https://creativecommons.org/licenses/by-nc/4.0/)

**Links:** [Publisher's page](https://aclanthology.org/2022.osact-1.4/) · [PDF](https://aclanthology.org/2022.osact-1.4.pdf)

**Citations:** [2](https://md.research.elamur.ai/free/en/papers/thubaity-2022-aranpcc-arabic-newspaper-covid-19/cited-by/) in our library, of the 149,439 papers whose bibliographies we read

## Summary

Large Arabic newspaper corpora existed before the pandemic, but none covered the COVID-19 period. The authors built AraNPCC to fill that gap. They crawled 88 widely read newspapers in 12 Arab countries from January 2019 to December 2021. Article pages were parsed with beautifulsoup4 to extract the text, title, topic and publication date. They then removed duplicates and normalised dates. Each newspaper's data is stored as yearly CSV files, with every article also saved as a separate text file. The final resource holds over 2 billion tokens across about 7.2 million articles and is released for non-commercial research.

Two case studies show how the corpus can be used:

- **Term frequency vs. case counts.** Monthly per-million frequency of the words "COVID" and "Corona" correlates positively with officially confirmed case counts, though the strength differs between countries.
- **Keyword comparison.** Top keywords for Algeria and Saudi Arabia are dominated by the pandemic in 2020, while the 2021 lists suggest the two countries were adapting.

The corpus is useful for updating Arabic language models and for studying language change over time.

## Key points

- AraNPCC contains 7,277,525 texts and 2,034,666,405 tokens from 88 newspapers in 12 countries. Egypt is the largest part, with about 2.9M texts. Some countries, such as the UAE, Lebanon, Qatar and Syria, could not be crawled.
- Newspapers were selected on three conditions: they had to be reliable, have an archive back to 2019, and number their articles with incremental IDs that make crawling easy.
- The raw data had 43 topic labels. These were merged into 13 classes (Society, Economy, Sports, Health and others), with No_Class used when no topic was found. Apart from deduplication and date normalisation, the texts are left unchanged.
- In the correlation study, Saudi Arabia ranks highest on all three measures (Pearson 0.577, Kendall 0.688, Spearman 0.857). Iraq's Pearson score is nearly zero (0.011). Kendall and Spearman are consistently higher than Pearson.
- Tokenisation used NLTK word_tokenize, case counts came from the CSSEGISandData/COVID-19 repository, and the corpus is hosted on archive.org.

## Interesting ideas
- The corpus deliberately includes one pre-pandemic year (2019) as a baseline, so any shift after the event can be measured against a normal period.
- The authors report Pearson, Kendall and Spearman side by side because corpus frequencies often violate one measure's distributional assumptions, and the coefficients do disagree noticeably.
- Noisy source labels (43 topic names, including near-duplicates such as Art/Arts) are mapped onto a smaller canonical set while the original labels stay intact.
- Text frequency is checked against an independent real-world time series (official case counts) to confirm that the corpus actually tracks events.
- Each article is stored twice, once as a CSV row and once as a separate file whose name encodes the source, topic, date and serial number, so the data survives the loss of either copy.

## For Mova

How these ideas could serve Mova, our small open language model — its files, stages and tests:

- **I1** (learn): Evaluation sets could pair texts that contain a phenomenon with comparable texts that lack it, acting as negative controls that expose detectors firing on everything.
- **I2** (learn): Agreement between witness parsers or teacher tiers could be reported with both rank-based and linear statistics, so one skewed metric cannot hide a weakness.
- **I3** (world): Ontology imports from Wikidata/FrameNet could keep the source label next to the canonical class, so mappings stay explainable and reversible.
- **I4** (world): The event order and entities that Mova extracts from news-like text could be checked against knowledge-graph dates, turning external facts into a sanity test.
- **I5** (learn): Silver training files could carry provenance in their names and keep a redundant copy, which makes it simple to trace and rebuild any example.

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

📚 Shelves (2): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 16,667 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 16,194 works

🧭 Topics (1): [Semitic languages](https://md.research.elamur.ai/free/en/topics/semitic-languages/) · 2,908 works

## Cite

Cite as: Abdulmohsen Al-Thubaity, Sakhar Alkhereyf and Alia O. Bahanshal. 2022. AraNPCC: The Arabic Newspaper COVID-19 Corpus. Proceedinsg of the 5th Workshop on Open-Source Arabic Corpora and Processing Tools with Shared Tasks on Qur'an QA and Fine-Grained Hate Speech Detection.

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

```bibtex
@inproceedings{althubaity2022aranpcc,
  title = {{AraNPCC: The Arabic Newspaper COVID-19 Corpus}},
  author = {Abdulmohsen Al-Thubaity, Sakhar Alkhereyf and Alia O. Bahanshal},
  booktitle = {Proceedinsg of the 5th Workshop on Open-Source Arabic Corpora and Processing Tools with Shared Tasks on Qur'an QA and Fine-Grained Hate Speech Detection},
  year = {2022},
  url = {https://aclanthology.org/2022.osact-1.4/}
}
```

- [← previous An Analysis of Dialogue Act Sequence Similarity Across Multiple Domains](https://md.research.elamur.ai/free/en/papers/enayet-2022-analysis-dialogue-act-sequence-similarity/)
- [next → Automatic Lemmatization of Old English Class III Strong Verbs (L-Y) with ALOEV3](https://md.research.elamur.ai/free/en/papers/alonso-2022-automatic-lemmatization-old-english-class/)
- [↑ back to papers of 2022](https://md.research.elamur.ai/free/en/papers/2022/page/8/)

## References

31 references, 6 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Al-Thubaity, A. O. (2015). A 700m+ Arabic corpus: KACST Arabic corpus design and construction. Language Resources and Evaluation, 49(3):721– 751. [↗](https://doi.org/10.1007/s10579-014-9284-1)
2. Amiruddin, N., Yassi, A. H., and Sukmawaty, S. (2022). Covid-19 blends: A new phenomenon in English Neologisms. Journal of Language and Linguistic Studies, 17(3). [find](https://scholar.google.com/scholar?q=Amiruddin%2C+N%2E%2C+Yassi%2C+A%2E+H%2E%2C+and+Sukmawaty%2C+S%2E+%282022%29%2E+Covid%2D19+blends%3A+A+new+phenomenon+in+English+Neologisms%2E+Journal+of+Language+and+Linguistic+Studies%2C+17%283%29%2E)
3. Antoun, W., Baly, F., and Hajj, H. (2020). [Arabert: Transformer-based model for arabic language understanding](https://md.research.elamur.ai/free/en/papers/antoun-2020-arabert-transformer-based-model-arabic/). arXiv preprint arXiv:2003.00104. [↗](https://arxiv.org/abs/2003.00104)
4. Arabzadeh, N., Khodabakhsh, M., and Bagheri, E. (2021). Bert-qpp: Contextualized pre-trained transformers for query performance prediction. In Proceedings of the 30th ACM International Conference on Information & Knowledge Management, pages 2857–2861. [find](https://scholar.google.com/scholar?q=Arabzadeh%2C+N%2E%2C+Khodabakhsh%2C+M%2E%2C+and+Bagheri%2C+E%2E+%282021%29%2E+Bert%2Dqpp%3A+Contextualized+pre%2Dtrained+transformers+for+query+performance+prediction%2E+In+Proceedings+of+the+30th+ACM+International+Conference+on+I)
5. Baker, P. and Heritage, F. (2021). How to use corpus linguistics in sociolinguistics: A case study of modal verb use, age and change over time. [↗](https://doi.org/10.4324/9780367076399-39)
6. Baron, A., Rayson, P., and Archer, D. (2009). Word frequency and key word statistics in corpus linguistics. Anglistik, 20(1):41–67. [find](https://scholar.google.com/scholar?q=Baron%2C+A%2E%2C+Rayson%2C+P%2E%2C+and+Archer%2C+D%2E+%282009%29%2E+Word+frequency+and+key+word+statistics+in+corpus+linguistics%2E+Anglistik%2C+20%281%29%3A41%E2%80%9367%2E)
7. Behpour, S., Mohammadi, M., Albert, M. V., Alam, Z. S., Wang, L., and Xiao, T. (2021). Automatic trend detection: Time-biased document clustering. Knowledge-Based Systems, 220:106907. [↗](https://doi.org/10.1016/j.knosys.2021.106907)
8. Belinkov, Y., Habash, N., Kilgarriff, A., Ordan, N., Roth, R., Suchomel, V., et al. (2013). arTenTen: a new, vast corpus for Arabic. Proceedings of WACL, 20. [find](https://scholar.google.com/scholar?q=Belinkov%2C+Y%2E%2C+Habash%2C+N%2E%2C+Kilgarriff%2C+A%2E%2C+Ordan%2C+N%2E%2C+Roth%2C+R%2E%2C+Suchomel%2C+V%2E%2C+et+al%2E+%282013%29%2E+arTenTen%3A+a+new%2C+vast+corpus+for+Arabic%2E+Proceedings+of+WACL%2C+20%2E)
9. Bird, S., Klein, E., and Loper, E. (2009). Natural language processing with Python: analyzing text with the natural language toolkit. " O’Reilly Media, Inc.". [find](https://scholar.google.com/scholar?q=Bird%2C+S%2E%2C+Klein%2C+E%2E%2C+and+Loper%2C+E%2E+%282009%29%2E+Natural+language+processing+with+Python%3A+analyzing+text+with+the+natural+language+toolkit%2E+%22+O%E2%80%99Reilly+Media%2C+Inc%2E%22%2E)
10. Curiac, C.-D., Banias, O., and Micea, M. (2022). Evaluating research trends from journal paper metadata, considering the research publication latency. Mathematics, 10(2):233. https://archive.org/details/AraNPCC [↗](https://doi.org/10.3390/math10020233)
11. Davies, M. (2021). The Coronavirus corpus: Design, construction, and use. International Journal of Corpus Linguistics. [find](https://scholar.google.com/scholar?q=Davies%2C+M%2E+%282021%29%2E+The+Coronavirus+corpus%3A+Design%2C+construction%2C+and+use%2E+International+Journal+of+Corpus+Linguistics%2E)
12. de Melo, T. and Figueiredo, C. M. (2020). A first public dataset from Brazilian twitter and news on COVID-19 in portuguese. Data in brief, 32:106179. [find](https://scholar.google.com/scholar?q=de+Melo%2C+T%2E+and+Figueiredo%2C+C%2E+M%2E+%282020%29%2E+A+first+public+dataset+from+Brazilian+twitter+and+news+on+COVID%2D19+in+portuguese%2E+Data+in+brief%2C+32%3A106179%2E)
13. Deuber, D., Hackert, S., Hänsel, E. C., Laube, A., Hejrani, M., and Laliberté, C. (2021). The norm orientation of English in the Caribbeana comparative study of newspaper writing from ten countries. American Speech, pages 1–40. [find](https://scholar.google.com/scholar?q=Deuber%2C+D%2E%2C+Hackert%2C+S%2E%2C+H%C3%A4nsel%2C+E%2E+C%2E%2C+Laube%2C+A%2E%2C+Hejrani%2C+M%2E%2C+and+Lalibert%C3%A9%2C+C%2E+%282021%29%2E+The+norm+orientation+of+English+in+the+Caribbeana+comparative+study+of+newspaper+writing+from+ten+countries%2E+A)
14. Devlin, J., Chang, M.-W., Lee, K., and Toutanova, K. (2018). [Bert: Pre-training of deep bidirectional transformers for language understanding](https://md.research.elamur.ai/free/en/papers/devlin-2019-bert-pre-training-deep-bidirectional/). arXiv preprint arXiv:1810.04805. [↗](https://arxiv.org/abs/1810.04805)
15. Dong, E., Du, H., and Gardner, L. (2020). An interactive web-based dashboard to track COVID-19 in real time. The Lancet infectious diseases, 20(5):533– 534. [↗](https://doi.org/10.1016/s1473-3099(20)30120-1)
16. El-Khair, I. A. (2016). [1.5 billion words Arabic corpus](https://md.research.elamur.ai/free/en/papers/el-khair-2016-1-5-billion-words-arabic/). arXiv preprint arXiv:1611.04033. [↗](https://arxiv.org/abs/1611.04033)
17. Gries, S. T. (2010). Useful statistics for corpus linguistics. A mosaic of corpus linguistics: Selected approaches, 66:269–291. [find](https://scholar.google.com/scholar?q=Gries%2C+S%2E+T%2E+%282010%29%2E+Useful+statistics+for+corpus+linguistics%2E+A+mosaic+of+corpus+linguistics%3A+Selected+approaches%2C+66%3A269%E2%80%93291%2E)
18. Hebbar, S. and Xie, Y. (2021). Covidbert-biomedical relation extraction for Covid-19. In The International FLAIRS Conference Proceedings, volume 34. [find](https://scholar.google.com/scholar?q=Hebbar%2C+S%2E+and+Xie%2C+Y%2E+%282021%29%2E+Covidbert%2Dbiomedical+relation+extraction+for+Covid%2D19%2E+In+The+International+FLAIRS+Conference+Proceedings%2C+volume+34%2E)
19. Imran, H. and Sharan, A. (2010). Co-occurrence based predictors for estimating query difficulty. In 2010 IEEE International Conference on Data Mining Workshops, pages 867–874. IEEE. [find](https://scholar.google.com/scholar?q=Imran%2C+H%2E+and+Sharan%2C+A%2E+%282010%29%2E+Co%2Doccurrence+based+predictors+for+estimating+query+difficulty%2E+In+2010+IEEE+International+Conference+on+Data+Mining+Workshops%2C+pages+867%E2%80%93874%2E+IEEE%2E)
20. Inoue, G., Alhafni, B., Baimukan, N., Bouamor, H., and Habash, N. (2021). [The interplay of variant, size, and task type in arabic pre-trained language models](https://md.research.elamur.ai/free/en/papers/inoue-2021-interplay-variant-size-task-type/). arXiv preprint arXiv:2103.06678. [↗](https://arxiv.org/abs/2103.06678)
21. Jakubíček, M., Kilgarriff, A., Kovář, V., Rychlỳ, P., and Suchomel, V. (2013). The TenTen corpus family. In 7th International Corpus Linguistics Conference CL, pages 125–127. [find](https://scholar.google.com/scholar?q=Jakub%C3%ADc%CC%8Cek%2C+M%2E%2C+Kilgarriff%2C+A%2E%2C+Kov%C3%A1r%CC%8C%2C+V%2E%2C+Rychly%CC%80%2C+P%2E%2C+and+Suchomel%2C+V%2E+%282013%29%2E+The+TenTen+corpus+family%2E+In+7th+International+Corpus+Linguistics+Conference+CL%2C+pages+125%E2%80%93127%2E)
22. Kilgarriff, A., Rychly, P., Smrz, P., and Tugwell, D. (2004). Itri-04-08 The Sketch Engine. Information Technology, 105(116). [find](https://scholar.google.com/scholar?q=Kilgarriff%2C+A%2E%2C+Rychly%2C+P%2E%2C+Smrz%2C+P%2E%2C+and+Tugwell%2C+D%2E+%282004%29%2E+Itri%2D04%2D08+The+Sketch+Engine%2E+Information+Technology%2C+105%28116%29%2E)
23. Kutuzov, A., Touileb, S., Mæhlum, P., Enstad, T. R., and Wittemann, A. (2022). NorDiaChange: Diachronic semantic change dataset for Norwegian. arXiv preprint arXiv:2201.05123. [↗](https://arxiv.org/abs/2201.05123)
24. Müller, M., Salathé, M., and Kummervold, P. E. (2020). Covid-twitter-bert: A natural language processing model to analyse covid-19 content on twitter. arXiv preprint arXiv:2005.07503. [↗](https://arxiv.org/abs/2005.07503)
25. Radford, A., Wu, J., Child, R., Luan, D., Amodei, D., Sutskever, I., et al. (2019). Language models are unsupervised multitask learners. OpenAI blog, 1(8):9. [find](https://scholar.google.com/scholar?q=Radford%2C+A%2E%2C+Wu%2C+J%2E%2C+Child%2C+R%2E%2C+Luan%2C+D%2E%2C+Amodei%2C+D%2E%2C+Sutskever%2C+I%2E%2C+et+al%2E+%282019%29%2E+Language+models+are+unsupervised+multitask+learners%2E+OpenAI+blog%2C+1%288%29%3A9%2E)
26. Rosin, G. D. and Radinsky, K. (2022). Temporal attention for language models. arXiv preprint arXiv:2202.02093. [↗](https://arxiv.org/abs/2202.02093)
27. Sonowal, G. (2020). Detecting phishing sms based on multiple correlation algorithms. SN Computer Science, 1(6):1–9. [↗](https://doi.org/10.1007/s42979-020-00377-8)
28. Suárez, P. J. O., Romary, L., and Sagot, B. (2020). [A monolingual approach to contextualized word embeddings for mid-resource languages](https://md.research.elamur.ai/free/en/papers/suarez-2020-monolingual-approach-contextualized-word-embeddings/). arXiv preprint arXiv:2006.06202. [↗](https://arxiv.org/abs/2006.06202)
29. Suwaileh, R., Kutlu, M., Fathima, N., Elsayed, T., and Lease, M. (2016). ArabicWeb16: A new crawl for today’s Arabic web. In Proceedings of the 39th International ACM SIGIR conference on Research and Development in Information Retrieval, pages 673– 676. [find](https://scholar.google.com/scholar?q=Suwaileh%2C+R%2E%2C+Kutlu%2C+M%2E%2C+Fathima%2C+N%2E%2C+Elsayed%2C+T%2E%2C+and+Lease%2C+M%2E+%282016%29%2E+ArabicWeb16%3A+A+new+crawl+for+today%E2%80%99s+Arabic+web%2E+In+Proceedings+of+the+39th+International+ACM+SIGIR+conference+on+Research+and+)
30. Zeroual, I., Goldhahn, D., Eckart, T., and Lakhouaja, A. (2019). [OSIAN: Open source international Arabic news corpus-preparation and integration into the clarin-infrastructure](https://md.research.elamur.ai/free/en/papers/zeroual-2019-osian-open-source-international-arabic/). In Proceedings of the Fourth Arabic Natural Language Processing Workshop, pages 175–182. 8. [↗](https://doi.org/10.60692/vm65h-ach57)
31. Language Resource References Parker, Robert, et al. (2011). Arabic Gigaword Fifth Edition. Linguistic Data Consortium, ISLRN 494-144-988-211-3. [↗](https://doi.org/10.35111/p02g-rw14)

*Updated 1 November 2026*