---
title: "ASAD: A Twitter-based Benchmark Arabic Sentiment Analysis Dataset — Alharbi 2020"
description: "Arabic sentiment analysis has lagged behind English and Chinese. 32 references, cited by 14 papers here."
authors: ["Basma Alharbi", "Hind Alamro", "Manal Alshehri", "Zuhair Khayyat", "Manal Kalkatawi", "Inji Ibrahim Jaber", "Xiangliang Zhang"]
year: 2020
venue: "arXiv:2011.00578"
source: "https://arxiv.org/abs/2011.00578"
doi: "10.48550/arxiv.2011.00578"
arxiv: "2011.00578"
licence: "arXiv — non-exclusive license to distribute"
licence_url: "http://arxiv.org/licenses/nonexclusive-distrib/1.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/alharbi-2020-asad-twitter-based-benchmark-arabic/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/alharbi-2020-asad-twitter-based-benchmark-arabic/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2020](https://md.research.elamur.ai/free/en/papers/2020/page/4/) / ASAD: A Twitter-based Benchmark Arabic Sentiment Analysis Dataset

# ASAD: A Twitter-based Benchmark Arabic Sentiment Analysis Dataset

**Authors:** [Basma Alharbi](https://md.research.elamur.ai/free/en/authors/basma-alharbi/), [Hind Alamro](https://md.research.elamur.ai/free/en/authors/hind-alamro/), [Manal Alshehri](https://md.research.elamur.ai/free/en/authors/manal-alshehri/), [Zuhair Khayyat](https://md.research.elamur.ai/free/en/authors/zuhair-khayyat/), [Manal Kalkatawi](https://md.research.elamur.ai/free/en/authors/manal-kalkatawi/), [Inji Ibrahim Jaber](https://md.research.elamur.ai/free/en/authors/inji-ibrahim-jaber/) and [Xiangliang Zhang](https://md.research.elamur.ai/free/en/authors/xiangliang-zhang/) · **Year:** 2020 · **Venue:** arXiv:2011.00578 · **Licence:** [arXiv — non-exclusive license to distribute](http://arxiv.org/licenses/nonexclusive-distrib/1.0/)

**Links:** [Publisher's page](https://arxiv.org/abs/2011.00578) · [PDF](https://arxiv.org/pdf/2011.00578) · [DOI 10.48550/arxiv.2011.00578](https://doi.org/10.48550/arxiv.2011.00578) · [arXiv 2011.00578](https://arxiv.org/abs/2011.00578) · [OpenAlex](https://openalex.org/W3103646405)

**Citations:** 25 citations (OpenAlex, 2026-10-14) · [14](https://md.research.elamur.ai/free/en/papers/alharbi-2020-asad-twitter-based-benchmark-arabic/cited-by/) in our library, of the 140,884 papers whose bibliographies we read

## Summary
Arabic sentiment analysis has lagged behind English and Chinese. One commonly cited reason is the shortage of large, carefully labelled benchmarks. Existing Twitter corpora are either big but labelled automatically from emojis (TEAD, ASTAD) or hand-labelled but small. The authors built ASAD, a three-class (positive/negative/neutral) tweet corpus labelled by humans. The tweets were sampled from a Twitter stream collected between 2012 and 2020 and filtered for length and media content. The Lucidya company annotated them in batches, with at least three people per tweet, and the corpus was cleaned to about 95K records. Agreement is moderate (Fleiss κ = 0.56). The authors also scored each annotator against the consensus label. ASAD underpins a KAUST competition, and the paper supplies reference baselines on two test sets: BOW and TF-IDF with logistic regression, BERT and AraBERT. AraBERT performs best. ASAD matters because it combines scale, manual labels and several dialects in one public resource.

## Key points
- Tweets were kept only if they were tagged "ar", had at least seven words (not counting hashtags and mentions), and contained no URLs, images, videos or blacklisted words. The 100,661 distinct tweets received 304,625 annotations and were cleaned down to 95,000.
- Most tweets come from Saudi Arabia (72%). By Lucidya's dialect models, the mix is 36% MSA, 31% Khaleeji, 22% Hijazi and 10% Egyptian.
- The corpus was labelled by 69 native-speaker annotators over 10 batches of 10K tweets each. Fleiss κ is 0.5035 for tweets with three annotators, 0.6067 for those with four, and 0.56 on average. Mean annotator accuracy against the final label is 0.85, and the top 10 annotators (over 60% of all labels) reach 0.88.
- AraBERT is the strongest baseline, with Macro-F1 0.68, Micro-F1 0.78 and F1-PN 0.59. TF-IDF-LR scores Macro-F1 0.62. Plain BERT recovers negatives poorly (negative recall about 0.28).
- The competition offered prizes of 10,000, 5,000 and 2,000 USD. Scores on TEST1 and TEST2 are nearly identical, so the public ranking is a reasonable guide to the final one.

## Interesting ideas
- After fixing final labels by majority vote (or an expert tie-break), each annotator is scored against them, which separates genuinely hard items from unreliable labellers.
- Annotation runs in batches, with a quality check after each one deciding who may continue. Quality control is thus spread across the project rather than done once at the end.
- Agreement is reported separately for items with three and with four annotators rather than as one opaque number, which shows how agreement shifts with labeller count.
- The headline metrics (average recall, F1 over positive and negative only) are designed so that the dominant neutral class cannot inflate the score.
- Two held-out test sets let the organisers confirm that the public leaderboard ranking predicts the private one.

## For Mova

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

- **I1** (learn): score every witness parser against the agreed silver arcs, so persistently deviant witnesses get down-weighted instead of treating all disagreement as item difficulty.
- **I2** (learn): put LLM teacher tiers through the same per-batch audit and stop routing data to any tier whose agreement with the checked subset drops.
- **I3** (learn): report silver-data agreement stratified by the number of agreeing witnesses, making it clear how trustworthy each stratum is.
- **I4** (answer): when most test items have a trivial majority answer (e.g. "no change" or "unknown"), report scores restricted to the informative answer classes so a lazy reader cannot look good.
- **I5** (learn): keep a second sealed evaluation split and check that rule or model rankings agree across both before accepting an improvement, as a guard against overfitting the dev set.

*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/en/shelves/linguistics/)

Topics: [😊 Sentiment & emotion](https://md.research.elamur.ai/free/en/topics/sentiment-emotion/) · [📏 Evaluation & benchmarks](https://md.research.elamur.ai/free/en/topics/evaluation-benchmarks/) · [🐪 Semitic languages](https://md.research.elamur.ai/free/en/topics/semitic-languages/)

## Institutions

- [King Abdullah University of Science and Technology](https://md.research.elamur.ai/free/en/universities/king-abdullah-university-of-science-and-technology/) — University, Jeddah, Saudi Arabia · 9 works · 856 citations
- [King Abdulaziz University](https://md.research.elamur.ai/free/en/universities/king-abdulaziz-university/) — University, Jeddah, Saudi Arabia · 12 works · 309 citations

## Cite

Cite as: Basma Alharbi, Hind Alamro, Manal Alshehri, Zuhair Khayyat, Manal Kalkatawi, Inji Ibrahim Jaber and Xiangliang Zhang. 2020. ASAD: A Twitter-based Benchmark Arabic Sentiment Analysis Dataset. arXiv:2011.00578.

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

```bibtex
@misc{alharbi2020asad,
  title = {{ASAD: A Twitter-based Benchmark Arabic Sentiment Analysis Dataset}},
  author = {Basma Alharbi, Hind Alamro, Manal Alshehri, Zuhair Khayyat, Manal Kalkatawi, Inji Ibrahim Jaber and Xiangliang Zhang},
  year = {2020},
  doi = {10.48550/arxiv.2011.00578},
  eprint = {2011.00578},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.48550/arxiv.2011.00578}
}
```

- [← previous The Hebrew of the Dead Sea Scrolls and the Tiberian Reading Tradition](https://md.research.elamur.ai/free/en/papers/hornkohl-2020-hebrew-dead-sea-scrolls-tiberian/)
- [next → Facilitating the Communication of Politeness through Fine-Grained Paraphrasing](https://md.research.elamur.ai/free/en/papers/fu-2020-facilitating-communication-politeness-through-fine/)
- [↑ back to papers of 2020](https://md.research.elamur.ai/free/en/papers/2020/page/4/)

## References

32 references, 8 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Imane Guellil, Houda Saâdane, Faical Azouaou, Billel Gueni, and Damien Nouvel. Arabic natural language processing: An overview. Journal of King Saud University-Computer and Information Sciences, 2019. [our card](https://md.research.elamur.ai/free/en/papers/guellil-2019-arabic-natural-language-processing-overview/)
2. Mahmoud Al-Ayyoub, Abed Allah Khamaiseh, Yaser Jararweh, and Mohammed N Al-Kabi. A comprehensive survey of arabic sentiment analysis. Information processing & management, 56(2):320–342, 2019. [find](https://scholar.google.com/scholar?q=Mahmoud+Al%2DAyyoub%2C+Abed+Allah+Khamaiseh%2C+Yaser+Jararweh%2C+and+Mohammed+N+Al%2DKabi%2E+A+comprehensive+survey+of+arabic+sentiment+analysis%2E+Information+processing+%26+management%2C+56%282%29%3A320%E2%80%93342%2C+2019%2E)
3. Noura Farra, Elie Challita, Rawad Abou Assi, and Hazem Hajj. Sentence-level and document-level sentiment mining for arabic texts. In 2010 IEEE international conference on data mining workshops, pages 1114–1119. IEEE, 2010. [find](https://scholar.google.com/scholar?q=Noura+Farra%2C+Elie+Challita%2C+Rawad+Abou+Assi%2C+and+Hazem+Hajj%2E+Sentence%2Dlevel+and+document%2Dlevel+sentiment+mining+for+arabic+texts%2E+In+2010+IEEE+international+conference+on+data+mining+workshops%2C+pages+)
4. Yann LeCun, Yoshua Bengio, and Geoffrey Hinton. Deep learning. nature, 521(7553):436–444, 2015. [find](https://scholar.google.com/scholar?q=Yann+LeCun%2C+Yoshua+Bengio%2C+and+Geoffrey+Hinton%2E+Deep+learning%2E+nature%2C+521%287553%29%3A436%E2%80%93444%2C+2015%2E)
5. Ahmad Al Sallab, Hazem Hajj, Gilbert Badaro, Ramy Baly, Wassim El-Hajj, and Khaled Shaban. Deep learning models for sentiment analysis in arabic. In Proceedings of the second workshop on Arabic natural language processing, pages 9–17, 2015. [find](https://scholar.google.com/scholar?q=Ahmad+Al+Sallab%2C+Hazem+Hajj%2C+Gilbert+Badaro%2C+Ramy+Baly%2C+Wassim+El%2DHajj%2C+and+Khaled+Shaban%2E+Deep+learning+models+for+sentiment+analysis+in+arabic%2E+In+Proceedings+of+the+second+workshop+on+Arabic+natura)
6. Sadam Al-Azani and El-Sayed M El-Alfy. Hybrid deep learning for sentiment polarity determination of arabic microblogs. In International Conference on Neural Information Processing, pages 491–500. Springer, 2017. BERT TEST1 0.62 0.39 0.85 0.62 0.74 0.60 0.28 0.92 0.60 0.50 AraBERT TEST1 TEST2 0.65 0.65 0.53 0.52 0.86 0.87 0.68 0.68 0.78 0.78 0.64 0.63 0.45 0.43 0.90 0.91 0.66 0.66 0.59 0.59 TEST2 0.61 0.38 0.85 0.62 0.74 0.59 0.27 0.92 0.59 0.50 [find](https://scholar.google.com/scholar?q=Sadam+Al%2DAzani+and+El%2DSayed+M+El%2DAlfy%2E+Hybrid+deep+learning+for+sentiment+polarity+determination+of+arabic+microblogs%2E+In+International+Conference+on+Neural+Information+Processing%2C+pages+491%E2%80%93500%2E+Spri)
7. Maha Heikal, Marwan Torki, and Nagwa El-Makky. Sentiment analysis of arabic tweets using deep learning. Procedia Computer Science, 142:114–122, 2018. [find](https://scholar.google.com/scholar?q=Maha+Heikal%2C+Marwan+Torki%2C+and+Nagwa+El%2DMakky%2E+Sentiment+analysis+of+arabic+tweets+using+deep+learning%2E+Procedia+Computer+Science%2C+142%3A114%E2%80%93122%2C+2018%2E)
8. Mohammad Al-Smadi, Omar Qawasmeh, Mahmoud Al-Ayyoub, Yaser Jararweh, and Brij Gupta. Deep recurrent neural network vs. support vector machine for aspect-based sentiment analysis of arabic hotels’ reviews. Journal of computational science, 27:386–393, 2018. [find](https://scholar.google.com/scholar?q=Mohammad+Al%2DSmadi%2C+Omar+Qawasmeh%2C+Mahmoud+Al%2DAyyoub%2C+Yaser+Jararweh%2C+and+Brij+Gupta%2E+Deep+recurrent+neural+network+vs%2E+support+vector+machine+for+aspect%2Dbased+sentiment+analysis+of+arabic+hotels%E2%80%99+revi)
9. Mahmoud Al-Ayyoub, Aya Nuseir, Kholoud Alsmearat, Yaser Jararweh, and Brij Gupta. Deep learning for arabic nlp: A survey. Journal of computational science, 26:522–531, 2018. [find](https://scholar.google.com/scholar?q=Mahmoud+Al%2DAyyoub%2C+Aya+Nuseir%2C+Kholoud+Alsmearat%2C+Yaser+Jararweh%2C+and+Brij+Gupta%2E+Deep+learning+for+arabic+nlp%3A+A+survey%2E+Journal+of+computational+science%2C+26%3A522%E2%80%93531%2C+2018%2E)
10. Ammar Mohammed and Rania Kora. Deep learning approaches for arabic sentiment analysis. Social Network Analysis and Mining, 9(1):52, 2019. [find](https://scholar.google.com/scholar?q=Ammar+Mohammed+and+Rania+Kora%2E+Deep+learning+approaches+for+arabic+sentiment+analysis%2E+Social+Network+Analysis+and+Mining%2C+9%281%29%3A52%2C+2019%2E)
11. Tomas Mikolov, Kai Chen, Greg Corrado, and Jeffrey Dean. Efficient estimation of word representations in vector space. arXiv preprint arXiv:1301.3781, 2013. [↗](https://arxiv.org/abs/1301.3781) [our card](https://md.research.elamur.ai/free/en/papers/mikolov-2013-efficient-estimation-word-representations-vector/)
12. Abu Bakr Soliman, Kareem Eissa, and Samhaa R El-Beltagy. Aravec: A set of arabic word embedding models for use in arabic nlp. Procedia Computer Science, 117:256–265, 2017. [find](https://scholar.google.com/scholar?q=Abu+Bakr+Soliman%2C+Kareem+Eissa%2C+and+Samhaa+R+El%2DBeltagy%2E+Aravec%3A+A+set+of+arabic+word+embedding+models+for+use+in+arabic+nlp%2E+Procedia+Computer+Science%2C+117%3A256%E2%80%93265%2C+2017%2E)
13. Obeida ElJundi, Wissam Antoun, Nour El Droubi, Hazem Hajj, Wassim El-Hajj, and Khaled Shaban. hulmona: The universal language model in arabic. In Proceedings of the Fourth Arabic Natural Language Processing Workshop, pages 68–77, 2019. [find](https://scholar.google.com/scholar?q=Obeida+ElJundi%2C+Wissam+Antoun%2C+Nour+El+Droubi%2C+Hazem+Hajj%2C+Wassim+El%2DHajj%2C+and+Khaled+Shaban%2E+hulmona%3A+The+universal+language+model+in+arabic%2E+In+Proceedings+of+the+Fourth+Arabic+Natural+Language+Proc)
14. Mahmoud Nabil, Mohamed Aly, and Amir Atiya. Labr: A large scale arabic sentiment analysis benchmark. arXiv preprint arXiv:1411.6718, 2014. [↗](https://arxiv.org/abs/1411.6718)
15. Muhammad Abdul-Mageed and Mona T Diab. Sana: A large scale multi-genre, multi-dialect lexicon for arabic subjectivity and sentiment analysis. In LREC, pages 1162–1169, 2014. [find](https://scholar.google.com/scholar?q=Muhammad+Abdul%2DMageed+and+Mona+T+Diab%2E+Sana%3A+A+large+scale+multi%2Dgenre%2C+multi%2Ddialect+lexicon+for+arabic+subjectivity+and+sentiment+analysis%2E+In+LREC%2C+pages+1162%E2%80%931169%2C+2014%2E)
16. Mohamed Aly and Amir Atiya. Labr: A large scale arabic book reviews dataset. In Proceedings of the 51st Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), pages 494–498, 2013. [find](https://scholar.google.com/scholar?q=Mohamed+Aly+and+Amir+Atiya%2E+Labr%3A+A+large+scale+arabic+book+reviews+dataset%2E+In+Proceedings+of+the+51st+Annual+Meeting+of+the+Association+for+Computational+Linguistics+%28Volume+2%3A+Short+Papers%29%2C+pages+)
17. Hady ElSahar and Samhaa R El-Beltagy. Building large arabic multi-domain resources for sentiment analysis. In International Conference on Intelligent Text Processing and Computational Linguistics, pages 23–34. Springer, 2015. [find](https://scholar.google.com/scholar?q=Hady+ElSahar+and+Samhaa+R+El%2DBeltagy%2E+Building+large+arabic+multi%2Ddomain+resources+for+sentiment+analysis%2E+In+International+Conference+on+Intelligent+Text+Processing+and+Computational+Linguistics%2C+pag)
18. Muhammad Abdul-Mageed and Mona T Diab. Awatif: A multi-genre corpus for modern standard arabic subjectivity and sentiment analysis. In LREC, volume 515, pages 3907–3914, 2012. [find](https://scholar.google.com/scholar?q=Muhammad+Abdul%2DMageed+and+Mona+T+Diab%2E+Awatif%3A+A+multi%2Dgenre+corpus+for+modern+standard+arabic+subjectivity+and+sentiment+analysis%2E+In+LREC%2C+volume+515%2C+pages+3907%E2%80%933914%2C+2012%2E)
19. Joseph L Fleiss. Measuring nominal scale agreement among many raters. Psychological bulletin, 76(5):378, 1971. [find](https://scholar.google.com/scholar?q=Joseph+L+Fleiss%2E+Measuring+nominal+scale+agreement+among+many+raters%2E+Psychological+bulletin%2C+76%285%29%3A378%2C+1971%2E)
20. Houssem Abdellaoui and Mounir Zrigui. Using tweets and emojis to build tead: an arabic dataset for sentiment analysis. Computación y Sistemas, 22(3):777–786, 2018. [find](https://scholar.google.com/scholar?q=Houssem+Abdellaoui+and+Mounir+Zrigui%2E+Using+tweets+and+emojis+to+build+tead%3A+an+arabic+dataset+for+sentiment+analysis%2E+Computaci%C3%B3n+y+Sistemas%2C+22%283%29%3A777%E2%80%93786%2C+2018%2E)
21. Kathrein Abu Kwaik, Stergios Chatzikyriakidis, Simon Dobnik, Motaz Saad, and Richard Johansson. An arabic tweets sentiment analysis dataset (atsad) using distant supervision and self training. In Proceedings of the 4th Workshop on Open-Source Arabic Corpora and Processing Tools, with a Shared Task on Offensive Language Detection, pages 1–8, 2020. [find](https://scholar.google.com/scholar?q=Kathrein+Abu+Kwaik%2C+Stergios+Chatzikyriakidis%2C+Simon+Dobnik%2C+Motaz+Saad%2C+and+Richard+Johansson%2E+An+arabic+tweets+sentiment+analysis+dataset+%28atsad%29+using+distant+supervision+and+self+training%2E+In+Proc)
22. Nora Al-Twairesh, Hend Al-Khalifa, AbdulMalik Al-Salman, and Yousef Al-Ohali. Arasenti-tweet: A corpus for arabic sentiment analysis of saudi tweets. Procedia Computer Science, 117:63–72, 2017. [find](https://scholar.google.com/scholar?q=Nora+Al%2DTwairesh%2C+Hend+Al%2DKhalifa%2C+AbdulMalik+Al%2DSalman%2C+and+Yousef+Al%2DOhali%2E+Arasenti%2Dtweet%3A+A+corpus+for+arabic+sentiment+analysis+of+saudi+tweets%2E+Procedia+Computer+Science%2C+117%3A63%E2%80%9372%2C+2017%2E)
23. Eshrag Refaee and Verena Rieser. An arabic twitter corpus for subjectivity and sentiment analysis. In LREC, pages 2268–2273, 2014. [find](https://scholar.google.com/scholar?q=Eshrag+Refaee+and+Verena+Rieser%2E+An+arabic+twitter+corpus+for+subjectivity+and+sentiment+analysis%2E+In+LREC%2C+pages+2268%E2%80%932273%2C+2014%2E)
24. Sara Rosenthal, Noura Farra, and Preslav Nakov. Semeval-2017 task 4: Sentiment analysis in twitter. In Proceedings of the 11th international workshop on semantic evaluation (SemEval-2017), pages 502–518, 2017. [our card](https://md.research.elamur.ai/free/en/papers/rosenthal-2017-semeval-2017-task-4-sentiment/)
25. Mahmoud Nabil, Mohamed Aly, and Amir Atiya. Astd: Arabic sentiment tweets dataset. In Proceedings of the 2015 conference on empirical methods in natural language processing, pages 2515–2519, 2015. [our card](https://md.research.elamur.ai/free/en/papers/nabil-2015-astd-arabic-sentiment-tweets-dataset/)
26. Saif M. Mohammad, Felipe Bravo-Marquez, Mohammad Salameh, and Svetlana Kiritchenko. Semeval-2018 Task 1: Affect in tweets. In Proceedings of International Workshop on Semantic Evaluation (SemEval-2018), New Orleans, LA, USA, 2018. [our card](https://md.research.elamur.ai/free/en/papers/mohammad-2018-semeval-2018-task-1-affect/)
27. Victoria Bobicev and Marina Sokolova. Inter-annotator agreement in sentiment analysis: Machine learning perspective. In RANLP, pages 97–102, 2017. [find](https://scholar.google.com/scholar?q=Victoria+Bobicev+and+Marina+Sokolova%2E+Inter%2Dannotator+agreement+in+sentiment+analysis%3A+Machine+learning+perspective%2E+In+RANLP%2C+pages+97%E2%80%93102%2C+2017%2E)
28. Fabrizio Sebastiani. An axiomatically derived measure for the evaluation of classification algorithms. In Proceedings of the 2015 International Conference on The Theory of Information Retrieval, pages 11–20, 2015. [find](https://scholar.google.com/scholar?q=Fabrizio+Sebastiani%2E+An+axiomatically+derived+measure+for+the+evaluation+of+classification+algorithms%2E+In+Proceedings+of+the+2015+International+Conference+on+The+Theory+of+Information+Retrieval%2C+pages)
29. Preslav Nakov, Alan Ritter, Sara Rosenthal, Fabrizio Sebastiani, and Veselin Stoyanov. Semeval-2016 task 4: Sentiment analysis in twitter. arXiv preprint arXiv:1912.01973, 2019. [↗](https://arxiv.org/abs/1912.01973) [our card](https://md.research.elamur.ai/free/en/papers/nakov-2016-semeval-2016-task-4-sentiment/)
30. Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. Bert: Pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:1810.04805, 2018. [↗](https://arxiv.org/abs/1810.04805) [our card](https://md.research.elamur.ai/free/en/papers/devlin-2019-bert-pre-training-deep-bidirectional/)
31. Wissam Antoun, Fady Baly, and Hazem Hajj. Arabert: Transformer-based model for arabic language understanding, 2020. [our card](https://md.research.elamur.ai/free/en/papers/antoun-2020-arabert-transformer-based-model-arabic/)
32. Qiang Yang, Hind Alamro, Somayah Albaradei, Adil Salhi, Xiaoting Lv, Changsheng Ma, Manal Alshehri, Inji Jaber, Faroug Tifratene, Wei Wang, Takashi Gojobori, Carlos M. Duarte, Xin Gao, and Xiangliang Zhang. SenWave: Monitoring the global sentiments under the covid-19 pandemic. arXiv preprint arXiv:2006.10842, 2020. [↗](https://arxiv.org/abs/2006.10842)