---
title: "AraSenTi: Large-Scale Twitter-Specific Arabic Sentiment Lexicons"
description: "Arabic sentiment analysis of Twitter lacks lexicons that reflect the informal, dialect-heavy and… 35 references, cited by 16 papers here."
authors: ["Nora Al-Twairesh", "Hend Al-Khalifa", "Abdulmalik Al-Salman"]
year: 2016
venue: "Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)"
source: "https://aclanthology.org/P16-1066/"
doi: "10.18653/v1/p16-1066"
licence: "CC BY 4.0 (ACL Anthology)"
licence_url: "https://creativecommons.org/licenses/by/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/twairesh-2016-arasenti-large-scale-twitter-specific/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/twairesh-2016-arasenti-large-scale-twitter-specific/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2016](https://md.research.elamur.ai/free/en/papers/2016/page/3/) / AraSenTi: Large-Scale Twitter-Specific Arabic Sentiment Lexicons

# AraSenTi: Large-Scale Twitter-Specific Arabic Sentiment Lexicons

**Authors:** [Nora Al-Twairesh](https://md.research.elamur.ai/free/en/authors/nora-al-twairesh/), [Hend Al-Khalifa](https://md.research.elamur.ai/free/en/authors/hend-al-khalifa/) and [Abdulmalik Al-Salman](https://md.research.elamur.ai/free/en/authors/abdulmalik-al-salman/) · **Year:** 2016 · **Venue:** [Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)](https://md.research.elamur.ai/free/en/venues/acl/) ([ACL 2016](https://md.research.elamur.ai/free/en/venues/acl/2016/)) · **Publisher:** [Association for Computational Linguistics](https://md.research.elamur.ai/free/en/publishers/association-for-computational-linguistics/) · **Licence:** [CC BY 4.0 (ACL Anthology)](https://creativecommons.org/licenses/by/4.0/)

**Links:** [Publisher's page](https://aclanthology.org/P16-1066/) · [PDF](https://aclanthology.org/P16-1066.pdf) · [DOI 10.18653/v1/p16-1066](https://doi.org/10.18653/v1/p16-1066) · [OpenAlex](https://openalex.org/W2516105622)

**Citations:** 61 citations (OpenAlex, 2026-10-14) · [16](https://md.research.elamur.ai/free/en/papers/twairesh-2016-arasenti-large-scale-twitter-specific/cited-by/) in our library, of the 149,439 papers whose bibliographies we read

## Summary
Arabic sentiment analysis of Twitter lacks lexicons that reflect the informal, dialect-heavy and fast-changing language of tweets. Resources derived from MSA dictionaries or SentiWordNet mappings (ArSenL, SLSA) miss this vocabulary. The authors collected about 6.3 million Arabic tweets over three months and filtered them to roughly 2.2 million. They used two kinds of distant labels: emoticons, and positive or negative sentiment keywords. From this corpus they built two lexicons, AraSenTi-Trans, which draws on the English Bing Liu and MPQA lexicons, and AraSenTi-PMI, which scores words by PMI association. Both were tested with a deliberately simple lexicon-based classifier, with and without negation handling. Testing covered one internal dataset and two external ones, and both lexicons were compared against the Twitter-derived DAHL lexicon. The best F-score was 89.58% on the internal data and 63.1–64.7% on the external data. Both lexicons beat DAHL everywhere. The work shows that cheap, noisily labelled social media data can yield useful domain-specific polarity resources.

## Key points
- The corpora are EMO-TWEET, collected with ":)" and ":(" plus "lang:ar" in Nov–Dec 2015, and KEY-TWEET, collected with 10 positive and 10 negative Arabic keywords in Jan 2016. About 6.3M tweets were gathered and about 2.2M remained after cleaning.
- Sentiment hashtags returned too few Arabic results, which the authors attribute to cultural differences in hashtag use, so the same words were searched as plain keywords instead.
- Cleaning removed retweets and tweets with URLs or media, since these were mostly spam. It also removed tweets containing non-Arabic characters, because the "lang:ar" filter leaked other languages.
- Evaluation used the internal AraSenti-Tweet set and the external ASTD and RR sets, against the DAHL lexicon. AraSenTi-PMI was best on RR (F 64.70) and on the internal set (F 89.58). AraSenTi-Trans was best on ASTD (F 63.10).
- Negation handling improved results by 1–4% on all datasets. For the PMI lexicon, a decision threshold of 1 on the summed score worked better than 0 or 0.5.

## Interesting ideas
- Corpus-specific lexicons can be built from distant labels such as emoticons and seed keywords, scoring each word by PMI with the positive class minus PMI with the negative class; this captures slang and dialect that no dictionary lists.
- When the collection signal fails because of how a community actually writes (here, hashtags that Arabs rarely use), switch to plain keywords rather than assuming the signal is universal.
- Negation was handled asymmetrically: a negator cancels the count of a positive word but leaves a negative word's count untouched, because cancelling negatives hurt accuracy.
- Tuning the decision threshold above the "natural" zero point corrected the lexicon's bias toward one class.
- Lexicons were scored on an in-domain set and on external datasets, which exposed a large drop (about 89% vs about 64%).

## For Mova

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

- **I1** (world): Mova could induce polarity or evaluative values for informal English words from noisily labelled text, storing each score with its PMI provenance so the value stays explainable.
- **I2** (learn): when harvesting silver data, Mova should check per genre that its trigger patterns really fire, and fall back to looser triggers measured by yield.
- **I3** (frames): polarity flipping under negation in prag could be treated as asymmetric and calibrated on data, not as a blind sign inversion.
- **I4** (answer): Mova's readers could tune decision cut-offs on held-out data against negative controls instead of assuming a symmetric zero.
- **I5** (learn): every induced resource should also be evaluated on out-of-source data, so that in-corpus scores are not mistaken for generality.

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

## Institutions

- [King Saud University](https://md.research.elamur.ai/free/en/universities/king-saud-university/) — University, Riyadh, Saudi Arabia · 37 works · 527 citations

📚 Shelves (4): [🖥️ 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 · [💭 Psychology](https://md.research.elamur.ai/free/en/shelves/psychology/) · 1,552 works · [💭 Experimental and Cognitive Psychology](https://md.research.elamur.ai/free/en/shelves/psychology/experimental-and-cognitive-psychology/) · 870 works

🧭 Topics (2): [😊 Sentiment & emotion](https://md.research.elamur.ai/free/en/topics/sentiment-emotion/) · 972 works · [Semitic languages](https://md.research.elamur.ai/free/en/topics/semitic-languages/) · 2,908 works

## Cite

Cite as: Nora Al-Twairesh, Hend Al-Khalifa and Abdulmalik Al-Salman. 2016. AraSenTi: Large-Scale Twitter-Specific Arabic Sentiment Lexicons. Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers).

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

```bibtex
@inproceedings{altwairesh2016arasenti,
  title = {{AraSenTi: Large-Scale Twitter-Specific Arabic Sentiment Lexicons}},
  author = {Nora Al-Twairesh, Hend Al-Khalifa and Abdulmalik Al-Salman},
  booktitle = {Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},
  year = {2016},
  doi = {10.18653/v1/p16-1066},
  url = {https://doi.org/10.18653/v1/p16-1066}
}
```

- [← previous Neural Network Translation Models for Grammatical Error Correction](https://md.research.elamur.ai/free/en/papers/chollampatt-2016-neural-network-translation-models-grammatical/)
- [next → Predicting the Direction of Derivation in English Conversion](https://md.research.elamur.ai/free/en/papers/kisselew-2016-predicting-direction-derivation-english-conversion/)
- [↑ back to papers of 2016](https://md.research.elamur.ai/free/en/papers/2016/page/3/)

## References

35 references, 15 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Muhammad Abdul-Mageed and Mona Diab. 2014. [SANA: A Large Scale Multi-Genre, Multi-Dialect Lexicon for Arabic Subjectivity and Sentiment Analysis](https://md.research.elamur.ai/free/en/papers/mageed-2014-sana-large-scale-multi-genre/). In In Proceedings of the Language Resources and Evaluation Conference (LREC), Reykjavik, Iceland.
2. Stefano Baccianella, Andrea Esuli, and Fabrizio Sebastiani. 2010. [SentiWordNet 3.0: An Enhanced Lexical Resource for Sentiment Analysis and Opinion Mining](https://md.research.elamur.ai/free/en/papers/baccianella-2010-sentiwordnet-3-0-enhanced-lexical/). In LREC, volume 10, pages 2200–2204.
3. Gilbert Badaro, Ramy Baly, Hazem Hajj, Nizar Habash, and Wassim El-Hajj. 2014. [A large scale Arabic sentiment lexicon for Arabic opinion mining](https://md.research.elamur.ai/free/en/papers/badaro-2014-large-scale-arabic-sentiment-lexicon/). ANLP 2014:165.
4. Sasha Blair-Goldensohn, Kerry Hannan, Ryan McDonald, Tyler Neylon, George A Reis, and Jeff Reynar. 2008. Building a sentiment summarizer for local service reviews. In WWW Workshop on NLP in the Information Explosion Era, volume 14, pages 339–348. [find](https://scholar.google.com/scholar?q=Sasha+Blair%2DGoldensohn%2C+Kerry+Hannan%2C+Ryan+McDonald%2C+Tyler+Neylon%2C+George+A+Reis%2C+and+Jeff+Reynar%2E+2008%2E+Building+a+sentiment+summarizer+for+local+service+reviews%2E+In+WWW+Workshop+on+NLP+in+the+Inform)
5. Kenneth Ward Church and Patrick Hanks. 1990. [Word association norms, mutual information, and lexicography](https://md.research.elamur.ai/free/en/papers/church-1990-word-association-norms-mutual-information/). Computational linguistics, 16(1):22–29. [↗](https://doi.org/10.3115/981623.981633)
6. Kareem Darwish and Walid Magdy. 2014. Arabic Information Retrieval. Foundations and Trends in Information Retrieval, 7(4):239–342. [↗](https://doi.org/10.1561/1500000031)
7. Dmitry Davidov, Oren Tsur, and Ari Rappoport. 2010. [Enhanced sentiment learning using twitter hashtags and smileys](https://md.research.elamur.ai/free/en/papers/davidov-2010-enhanced-sentiment-learning-twitter-hashtags/). In Proceedings of the 23rd International Conference on Computational Linguistics: Posters, pages 241–249. Association for Computational Linguistics.
8. Mona Diab, Kadri Hacioglu, and Daniel Jurafsky. 2007. Automated methods for processing arabic text: from tokenization to base phrase chunking. [find](https://scholar.google.com/scholar?q=Mona+Diab%2C+Kadri+Hacioglu%2C+and+Daniel+Jurafsky%2E+2007%2E+Automated+methods+for+processing+arabic+text%3A+from+tokenization+to+base+phrase+chunking%2E)
9. Arabic Rehab M Duwairi. 2015. Sentiment analysis for dialectical Arabic. In 6th International Conference on Information and Communication Systems (ICICS), 2015, pages 166–170. IEEE. [find](https://scholar.google.com/scholar?q=Arabic+Rehab+M+Duwairi%2E+2015%2E+Sentiment+analysis+for+dialectical+Arabic%2E+In+6th+International+Conference+on+Information+and+Communication+Systems+%28ICICS%29%2C+2015%2C+pages+166%E2%80%93170%2E+IEEE%2E)
10. Andrea Esuli and Fabrizio Sebastiani. 2005. Determining the semantic orientation of terms through gloss classification. In Proceedings of the 14th ACM international conference on Information and knowledge management, pages 617–624. ACM. [find](https://scholar.google.com/scholar?q=Andrea+Esuli+and+Fabrizio+Sebastiani%2E+2005%2E+Determining+the+semantic+orientation+of+terms+through+gloss+classification%2E+In+Proceedings+of+the+14th+ACM+international+conference+on+Information+and+knowl)
11. Andrea Esuli and Fabrizio Sebastiani. 2006. [Sentiwordnet: A publicly available lexical resource for opinion mining](https://md.research.elamur.ai/free/en/papers/esuli-2006-sentiwordnet-publicly-available-lexical-resource/). In In Proceedings of the 5th Conference on Language Resources and Evaluation (LREC’06), volume 6, pages 417–422.
12. Alec Go, Richa Bhayani, and Lei Huang. 2009. Twitter sentiment classification using distant supervision. CS224N Project Report, Stanford:1–12. [find](https://scholar.google.com/scholar?q=Alec+Go%2C+Richa+Bhayani%2C+and+Lei+Huang%2E+2009%2E+Twitter+sentiment+classification+using+distant+supervision%2E+CS224N+Project+Report%2C+Stanford%3A1%E2%80%9312%2E)
13. Nizar Habash and Owen Rambow. 2005. [Arabic tokenization, part-of-speech tagging and morphological disambiguation in one fell swoop](https://md.research.elamur.ai/free/en/papers/habash-2005-arabic-tokenization-part-speech-tagging/). In Proceedings of the 43rd Annual Meeting on Association for Computational Linguistics, pages 573–580. Association for Computational Linguistics.
14. Nizar Habash, Owen Rambow, and Ryan Roth. 2009. Mada+ tokan: A toolkit for arabic tokenization, diacritization, morphological disambiguation, pos tagging, stemming and lemmatization. In Proceedings of the 2nd International Conference on Arabic Language Resources and Tools (MEDAR), Cairo, Egypt, pages 102–109. [find](https://scholar.google.com/scholar?q=Nizar+Habash%2C+Owen+Rambow%2C+and+Ryan+Roth%2E+2009%2E+Mada%2B+tokan%3A+A+toolkit+for+arabic+tokenization%2C+diacritization%2C+morphological+disambiguation%2C+pos+tagging%2C+stemming+and+lemmatization%2E+In+Proceedings+of)
15. Nizar Habash, Ryan Roth, Owen Rambow, Ramy Eskander, and Nadi Tomeh. 2013. [Morphological Analysis and Disambiguation for Dialectal Arabic](https://md.research.elamur.ai/free/en/papers/habash-2013-morphological-analysis-disambiguation-dialectal-arabic/). In HLT-NAACL, pages 426–432. Citeseer.
16. Nizar Y Habash. 2010. Introduction to Arabic natural language processing. Synthesis Lectures on Human Language Technologies, 3(1):1–187. [↗](https://doi.org/10.1007/978-3-031-02139-8)
17. Vasileios Hatzivassiloglou and Kathleen R McKeown. 1997. Predicting the semantic orientation of adjectives. In Proceedings of the 35th annual meeting of the association for computational linguistics and eighth conference of the european chapter of the association for computational linguistics, pages 174– 181. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Vasileios+Hatzivassiloglou+and+Kathleen+R+McKeown%2E+1997%2E+Predicting+the+semantic+orientation+of+adjectives%2E+In+Proceedings+of+the+35th+annual+meeting+of+the+association+for+computational+linguistics+a)
18. Minqing Hu and Bing Liu. 2004. Mining and summarizing customer reviews. In Proceedings of the tenth ACM SIGKDD international conference on Knowledge discovery and data mining, pages 168– 177. ACM. [find](https://scholar.google.com/scholar?q=Minqing+Hu+and+Bing+Liu%2E+2004%2E+Mining+and+summarizing+customer+reviews%2E+In+Proceedings+of+the+tenth+ACM+SIGKDD+international+conference+on+Knowledge+discovery+and+data+mining%2C+pages+168%E2%80%93+177%2E+ACM%2E)
19. Internet World Stats. 2015. Internet World Stats. November. [find](https://scholar.google.com/scholar?q=Internet+World+Stats%2E+2015%2E+Internet+World+Stats%2E+November%2E)
20. Jaap Kamps. 2004. Using wordnet to measure semantic orientations of adjectives. In Proceedings of the 4th International Conference on Language Resources and Evaluation (LREC 2004). [find](https://scholar.google.com/scholar?q=Jaap+Kamps%2E+2004%2E+Using+wordnet+to+measure+semantic+orientations+of+adjectives%2E+In+Proceedings+of+the+4th+International+Conference+on+Language+Resources+and+Evaluation+%28LREC+2004%29%2E)
21. Svetlana Kiritchenko, Xiaodan Zhu, and Saif M Mohammad. 2014. Sentiment analysis of short informal texts. Journal of Artificial Intelligence Research, 50:723–762. [↗](https://doi.org/10.1613/jair.4272)
22. Bing Liu. 2012. Sentiment analysis and opinion mining. Synthesis Lectures on Human Language Technologies, 5(1):1–167. [↗](https://doi.org/10.1007/978-3-031-02145-9)
23. Hanaa Mobarz, Mohsen Rashown, and Ibrahim Farag. 2014. Using Automated Lexical Resources in Arabic Sentence Subjectivity. International Journal of Artificial Intelligence & Applications, 5(6):1. [↗](https://doi.org/10.5121/ijaia.2014.5601)
24. Saif M Mohammad, Mohammad Salameh, and Svetlana Kiritchenko. 2015. How Translation Alters Sentiment. Journal of Artificial Intelligence Research, 54:1–20. [↗](https://doi.org/10.1613/jair.4787)
25. Mahmoud Nabil, Mohamed Aly, and Amir F Atiya. 2015. [ASTD: Arabic Sentiment Tweets Dataset](https://md.research.elamur.ai/free/en/papers/nabil-2015-astd-arabic-sentiment-tweets-dataset/). In Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing, pages 2515–2519. [↗](https://doi.org/10.60692/3ee2g-zjt73)
26. Alexander Pak and Patrick Paroubek. 2010. [Twitter as a Corpus for Sentiment Analysis and Opinion Mining](https://md.research.elamur.ai/free/en/papers/pak-2010-twitter-as-corpus-sentiment-analysis/). In Proceedings of the Sixth International Conference on Language Resources and Evaluation (LREC 2010), Valleta,Malta. European Language Resources Association (ELRA).
27. Arfath Pasha, Mohamed Al-Badrashiny, Ahmed El Kholy, Ramy Eskander, Mona Diab, Nizar Habash, Manoj Pooleery, Owen Rambow, and Ryan Roth. 2014. [Madamira: A fast, comprehensive tool for morphological analysis and disambiguation of arabic](https://md.research.elamur.ai/free/en/papers/pasha-2014-madamira-fast-comprehensive-tool-morphological/). In In Proceedings of the 9th International Conference on Language Resources and Evaluation, LREC 2014, Reykjavik, Iceland. European Language Resources Association (ELRA).
28. Delip Rao and Deepak Ravichandran. 2009. Semi-supervised polarity lexicon induction. In Proceedings of the 12th Conference of the European Chapter of the Association for Computational Linguistics, pages 675–682. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Delip+Rao+and+Deepak+Ravichandran%2E+2009%2E+Semi%2Dsupervised+polarity+lexicon+induction%2E+In+Proceedings+of+the+12th+Conference+of+the+European+Chapter+of+the+Association+for+Computational+Linguistics%2C+pag)
29. Eshrag Refaee and Verena Rieser. 2014. [An Arabic Twitter Corpus for Subjectivity and Sentiment Analysis](https://md.research.elamur.ai/free/en/papers/refaee-2014-arabic-twitter-corpus-subjectivity-sentiment/). In In Proceedings of the 9th International Conference on Language Resources and Evaluation, LREC 2014, Reykjavik, Iceland. European Language Resources Association (ELRA).
30. Ryan Roth, Owen Rambow, Nizar Habash, Mona Diab, and Cynthia Rudin. 2008. [Arabic morphological tagging, diacritization, and lemmatization using lexeme models and feature ranking](https://md.research.elamur.ai/free/en/papers/roth-2008-arabic-morphological-tagging-diacritization-lemmatization/). In Proceedings of the 46th Annual Meeting of the Association for Computational Linguistics on Human Language Technologies: Short Papers, pages 117–120. Association for Computational Linguistics.
31. Duyu Tang, Furu Wei, Bing Qin, Ming Zhou, and Ting Liu. 2014. Building Large-Scale Twitter-Specific Sentiment Lexicon: A Representation Learning Approach. In COLING, pages 172–182. [find](https://scholar.google.com/scholar?q=Duyu+Tang%2C+Furu+Wei%2C+Bing+Qin%2C+Ming+Zhou%2C+and+Ting+Liu%2E+2014%2E+Building+Large%2DScale+Twitter%2DSpecific+Sentiment+Lexicon%3A+A+Representation+Learning+Approach%2E+In+COLING%2C+pages+172%E2%80%93182%2E)
32. Peter D Turney. 2002. [Thumbs up or thumbs down?: semantic orientation applied to unsupervised classification of reviews](https://md.research.elamur.ai/free/en/papers/turney-2002-thumbs-up-or-thumbs-down/). In Proceedings of the 40th annual meeting on association for computational linguistics, pages 417–424. Association for Computational Linguistics.
33. Peter Turney and Michael L Littman. 2002. Unsupervised learning of semantic orientation from a hundred-billion-word corpus. Technical report, National Research Council Canada, NRC Institute for Information Technology; National Research Council Canada. [↗](https://doi.org/10.4224/8914027)
34. Gbolahan K Williams and Sarabjot Singh Anand. 2009. Predicting the Polarity Strength of Adjectives Using WordNet. In Third International AAAI Conference on Weblogs and Social Media. [find](https://scholar.google.com/scholar?q=Gbolahan+K+Williams+and+Sarabjot+Singh+Anand%2E+2009%2E+Predicting+the+Polarity+Strength+of+Adjectives+Using+WordNet%2E+In+Third+International+AAAI+Conference+on+Weblogs+and+Social+Media%2E)
35. Theresa Wilson, Janyce Wiebe, and Paul Hoffmann. 2005. [Recognizing contextual polarity in phrase-level sentiment analysis](https://md.research.elamur.ai/free/en/papers/wilson-2005-recognizing-contextual-polarity-phrase-level/). In Proceedings of the conference on human language technology and empirical methods in natural language processing, pages 347–354. Association for Computational Linguistics.

*Updated 1 November 2026*