---
title: "Subjectivity and Sentiment Analysis of Modern Standard Arabic"
description: "They build a cascaded classifier that first decides subjective vs. objective, then positive vs. negative. 23 references, cited by 37 papers here."
authors: ["Ahmed Mourad", "Kareem Darwish"]
year: 2013
venue: "Proceedings of the 4th Workshop on Computational Approaches to Subjectivity, Sentiment and Social Media Analysis"
source: "https://aclanthology.org/W13-1608/"
licence: "CC BY-NC-SA 3.0 (ACL Anthology, before 2016)"
licence_url: "https://creativecommons.org/licenses/by-nc-sa/3.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/mourad-2013-subjectivity-sentiment-analysis-modern-standard/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/mourad-2013-subjectivity-sentiment-analysis-modern-standard/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2013](https://md.research.elamur.ai/free/en/papers/2013/) / Subjectivity and Sentiment Analysis of Modern Standard Arabic and Arabic Microblogs

# Subjectivity and Sentiment Analysis of Modern Standard Arabic and Arabic Microblogs

**Authors:** Ahmed Mourad and [Kareem Darwish](https://md.research.elamur.ai/free/en/authors/kareem-darwish/) · **Year:** 2013 · **Venue:** [Proceedings of the 4th Workshop on Computational Approaches to Subjectivity, Sentiment and Social Media Analysis](https://md.research.elamur.ai/free/en/venues/naacl/) ([NAACL 2013](https://md.research.elamur.ai/free/en/venues/naacl/2013/)) · **Publisher:** [Association for Computational Linguistics](https://md.research.elamur.ai/free/en/publishers/association-for-computational-linguistics/) · **Licence:** [CC BY-NC-SA 3.0 (ACL Anthology, before 2016)](https://creativecommons.org/licenses/by-nc-sa/3.0/)

**Links:** [Publisher's page](https://aclanthology.org/W13-1608/) · [PDF](https://aclanthology.org/W13-1608.pdf) · [OpenAlex](https://openalex.org/W2162357159)

**Citations:** 151 citations (OpenAlex, 2026-10-14) · [37](https://md.research.elamur.ai/free/en/papers/mourad-2013-subjectivity-sentiment-analysis-modern-standard/cited-by/) in our library, of the 149,439 papers whose bibliographies we read

## Summary
The paper tackles subjectivity and sentiment analysis (SSA) for Arabic in two genres: Modern Standard Arabic (MSA) news and dialect-heavy Twitter posts. Arabic SSA has two gaps. Its sentiment lexicons are small, and no public labelled tweet data existed. The authors respond in three ways:
- They hand-annotate a new set of 2,300 Arabic tweets.
- They build a cascaded classifier that first decides subjective vs. objective, then positive vs. negative. Features include stems, POS tags, lexicon polarity and tweet-specific cues.
- They grow the manually curated ArabSenti lexicon through graph reinforcement, a random walk over Arabic-English MT phrase tables covering both MSA and dialect.

The baseline beats earlier published results for MSA news subjectivity. On tweets, training on in-domain data matters most. The expanded lexicon roughly quadruples lexicon matches and gives modest gains for tweet sentiment. On news it barely helps. The work shows how to bootstrap SSA resources for a morphologically rich, dialect-fragmented language.

## Key points
- The new corpus has 2,300 Arabic tweets. About 70% are subjective, and among the polar subjective tweets about 66% are positive. Roughly 13.5% of tweets were judged sarcastic.
- Two lexicons were compared: MPQA (about 8,000 English entries, machine-translated) and the manually built ArabSenti. Only ArabSenti was expanded, because translation errors in MPQA would likely have been amplified.
- For tweet subjectivity, simply predicting the majority class won. Even so, a classifier trained on tweets beat one trained on MSA news.
- Training the sentiment classifier on tweets instead of news raised accuracy by 17.4 points absolute. Positive-class F-measure rose by 17.2 points and negative-class by 16.1.
- Lexicon expansion changed the number of lexicon matches as follows:

| Data | Baseline | Expanded |
|---|---|---|
| News | 105k | 567k (English-MSA table) / 550k (English-Dialect table) |
| Tweet subjectivity | 19k | 75k |
| Tweet sentiment | 7k | 28k |

  Gains appeared only on tweets, at about 1–3 points for sentiment. The larger English-MSA phrase table beat the English-Dialect one.

## Interesting ideas
- Seed a lexicon with a small, trusted hand-made list and spread its labels by random walks through bilingual phrase tables, deliberately skipping the noisier machine-translated list so errors are not multiplied.
- Larger lexicon coverage did not reliably improve accuracy (fivefold more matches on news, almost no gain), so coverage and usefulness should be measured separately.
- A retweet marker or URL appearing outside its usual position was used as a cue that the poster added their own, likely opinionated, words to forwarded content.
- The majority-class baseline was kept as a serious competitor and turned out unbeatable for tweet subjectivity. The authors also warn that artificially class-balanced training data distorts priors.
- A feature was dropped (negative emoticons) because it appeared too rarely in training to be trusted, even though it seemed intuitively informative.

## For Mova

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

- **I1** (world): Verb-class and ontology seeds could be expanded through translation-pivot graphs over WordNet/Wikidata labels, with every added entry keeping the walk path that produced it as an explanation.
- **I2** (learn): When induced repair rules or knowledge entries are added, report coverage growth and task effect side by side, so that "fires more often" is not mistaken for "helps".
- **I3** (frames): Departures from conventional placement of quotes and attributions could help separate the author's own speech act from reported content when assigning opinion holders.
- **I4** (learn): Trivial predictors should run in every evaluation as negative controls, and evaluation sets should keep natural label distributions.
- **I5** (tokens): Surface cues such as emoticons or elongation in the tagger could be gated by minimum training frequency, with the reason for exclusion logged.

*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

- [Qatar Foundation](https://md.research.elamur.ai/free/en/ngos/qatar-foundation/) — Nonprofit, Doha, Qatar · 17 works · 2,111 citations

📚 Shelves (6): [🖥️ 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 · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 16,039 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 15,891 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: Ahmed Mourad and Kareem Darwish. 2013. Subjectivity and Sentiment Analysis of Modern Standard Arabic and Arabic Microblogs. Proceedings of the 4th Workshop on Computational Approaches to Subjectivity, Sentiment and Social Media Analysis.

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

```bibtex
@inproceedings{mourad2013subjectivity,
  title = {{Subjectivity and Sentiment Analysis of Modern Standard Arabic and Arabic Microblogs}},
  author = {Ahmed Mourad and Kareem Darwish},
  booktitle = {Proceedings of the 4th Workshop on Computational Approaches to Subjectivity, Sentiment and Social Media Analysis},
  year = {2013},
  url = {https://aclanthology.org/W13-1608/}
}
```

- [← previous Sentence Level Dialect Identification in Arabic](https://md.research.elamur.ai/free/en/papers/elfardy-2013-sentence-level-dialect-identification-arabic/)
- [next → Target Language Adaptation of Discriminative Transfer Parsers](https://md.research.elamur.ai/free/en/papers/tackstrom-2013-target-language-adaptation-discriminative-transfer/)
- [↑ back to papers of 2013](https://md.research.elamur.ai/free/en/papers/2013/)

## References

23 references, 11 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Ahmed Abbasi, Hsinchun Chen, and Arab Salem. 2008. Sentiment analysis in multiple languages: Feature selection for opinion classification in web forums. ACM Transactions on Information Systems (TOIS), 26(3):12, 2008. [find](https://scholar.google.com/scholar?q=Ahmed+Abbasi%2C+Hsinchun+Chen%2C+and+Arab+Salem%2E+2008%2E+Sentiment+analysis+in+multiple+languages%3A+Feature+selection+for+opinion+classification+in+web+forums%2E+ACM+Transactions+on+Information+Systems+%28TOIS%29%2C)
2. Muhammad Abdul-Mageed, Mona Diab, and Mohammed Korayem. 2011. [Subjectivity and sentiment analysis of modern standard arabic](https://md.research.elamur.ai/free/en/papers/mageed-2011-subjectivity-sentiment-analysis-modern-standard/). In Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies, pages 587–591, 2011.
3. Muhammad Abdul-Mageed and Mona T. Diab. 2011. [Subjectivity and sentiment annotation of modern standard arabic newswire](https://md.research.elamur.ai/free/en/papers/mageed-2011-subjectivity-sentiment-annotation-modern-standard/).
4. Muhammad Abdul-Mageed, Sandra Kübler, and Mona Diab. 2012. [Samar: A system for subjectivity and sentiment analysis of arabic social media](https://md.research.elamur.ai/free/en/papers/mageed-2012-samar-system-subjectivity-sentiment-analysis/). WASSA 2012, page 19, 2012.
5. Carmen Banea, Rada Mihalcea, Janyce Wiebe, and Samer Hassan. 2008. [Multilingual subjectivity analysis using machine translation](https://md.research.elamur.ai/free/en/papers/banea-2008-multilingual-subjectivity-analysis-machine-translation/). In Proceedings of the Conference on Empirical Methods in Natural Language Processing, pages 127–135.
6. Carmen Banea, Rada Mihalcea, and Janyce Wiebe. 2010. Multilingual subjectivity: are more languages better? In Proceedings of the 23rd International Conference on Computational Linguistics, pages 28–36. Association for Computational Linguistics, 2010. [find](https://scholar.google.com/scholar?q=Carmen+Banea%2C+Rada+Mihalcea%2C+and+Janyce+Wiebe%2E+2010%2E+Multilingual+subjectivity%3A+are+more+languages+better%3F+In+Proceedings+of+the+23rd+International+Conference+on+Computational+Linguistics%2C+pages+28%E2%80%9336)
7. Luciano Barbosa and Junlan Feng. 2010. [Robust sentiment detection on twitter from biased and noisy data](https://md.research.elamur.ai/free/en/papers/barbosa-2010-robust-sentiment-detection-twitter-biased/). In Proceedings of the 23rd International Conference on Computational Linguistics: Posters, pages 36–44. Association for Computational Linguistics.
8. Steven Bird. 2006. [NLTK: the natural language toolkit](https://md.research.elamur.ai/free/en/papers/bird-2006-nltk-natural-language-toolkit/). In Proceedings of the COLING/ACL on Interactive presentation sessions, pages 69–72. Association for Computational Linguistics, 2006.
9. Kareem Darwish, Walid Magdy, and Ahmed Mourad. 2012. Language processing for arabic microblog retrieval. In Proceedings of the 21st ACM international conference on Information and knowledge management, pages 2427–2430. [find](https://scholar.google.com/scholar?q=Kareem+Darwish%2C+Walid+Magdy%2C+and+Ahmed+Mourad%2E+2012%2E+Language+processing+for+arabic+microblog+retrieval%2E+In+Proceedings+of+the+21st+ACM+international+conference+on+Information+and+knowledge+management)
10. Mona Diab. 2009. Second Generation Tools (AMIRA 2.0): Fast and Robust Tokenization, POS tagging, and Base Phrase Chunking. In 2nd International Conference on Arabic Language Resources and Tools, 2009. [find](https://scholar.google.com/scholar?q=Mona+Diab%2E+2009%2E+Second+Generation+Tools+%28AMIRA+2%2E0%29%3A+Fast+and+Robust+Tokenization%2C+POS+tagging%2C+and+Base+Phrase+Chunking%2E+In+2nd+International+Conference+on+Arabic+Language+Resources+and+Tools%2C+2009%2E)
11. Ali El-Kahky, Kareem Darwish, Ahmed Saad Aldein, Mohamed Abd El-Wahab, Ahmed Hefny, and Waleed Ammar. 2011. Improved transliteration mining using graph reinforcement. In Proceedings of the Conference on Empirical Methods in Natural Language Processing, pages 1384–1393. Association for Computational Linguistics, 2011. [find](https://scholar.google.com/scholar?q=Ali+El%2DKahky%2C+Kareem+Darwish%2C+Ahmed+Saad+Aldein%2C+Mohamed+Abd+El%2DWahab%2C+Ahmed+Hefny%2C+and+Waleed+Ammar%2E+2011%2E+Improved+transliteration+mining+using+graph+reinforcement%2E+In+Proceedings+of+the+Conference+)
12. Spence Green and John DeNero. 2012. A Class-Based Agreement Model for Generating Accurately Inflected Translations. ACL 2012. [find](https://scholar.google.com/scholar?q=Spence+Green+and+John+DeNero%2E+2012%2E+A+Class%2DBased+Agreement+Model+for+Generating+Accurately+Inflected+Translations%2E+ACL+2012%2E)
13. Philipp Koehn, Hieu Hoang, Alexandra Birch, Chris Callison-Burch, Marcello Federico, Nicola Bertoldi, Brooke Cowan, Wade Shen, Christine Moran, Richard Zens, and others. 2007. Moses: Open source toolkit for statistical machine translation. In Annual meeting-association for computational linguistics. [find](https://scholar.google.com/scholar?q=Philipp+Koehn%2C+Hieu+Hoang%2C+Alexandra+Birch%2C+Chris+Callison%2DBurch%2C+Marcello+Federico%2C+Nicola+Bertoldi%2C+Brooke+Cowan%2C+Wade+Shen%2C+Christine+Moran%2C+Richard+Zens%2C+and+others%2E+2007%2E+Moses%3A+Open+source+toolk)
14. Stanley Kok and Chris Brockett. 2010. Hitting the right paraphrases in good time HLT-NAACL-2010, pages 145–153. [find](https://scholar.google.com/scholar?q=Stanley+Kok+and+Chris+Brockett%2E+2010%2E+Hitting+the+right+paraphrases+in+good+time+HLT%2DNAACL%2D2010%2C+pages+145%E2%80%93153%2E)
15. Alok Kothari, Walid Magdy, Kareem Darwish, Ahmed Mourad, and Ahmed Taei. 2013. Detecting Comments on News Articles in Microblogs ICWSM, pages 145–153. [find](https://scholar.google.com/scholar?q=Alok+Kothari%2C+Walid+Magdy%2C+Kareem+Darwish%2C+Ahmed+Mourad%2C+and+Ahmed+Taei%2E+2013%2E+Detecting+Comments+on+News+Articles+in+Microblogs+ICWSM%2C+pages+145%E2%80%93153%2E)
16. Bing Liu. 2010. Sentiment analysis and subjectivity. Handbook of Natural Language Processing, pages 627–666, 2010. [find](https://scholar.google.com/scholar?q=Bing+Liu%2E+2010%2E+Sentiment+analysis+and+subjectivity%2E+Handbook+of+Natural+Language+Processing%2C+pages+627%E2%80%93666%2C+2010%2E)
17. Rada Mihalcea, Carmen Banea, and Janyce Wiebe. 2007. [Learning multilingual subjective language via cross-lingual projections](https://md.research.elamur.ai/free/en/papers/mihalcea-2007-learning-multilingual-subjective-language-cross/). In ACL-2007, volume 45, page 976, 2007.
18. 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 LREC, volume 2010, 2010.
19. Janyce Wiebe and Ellen Riloff. 2005. Creating subjective and objective sentence classifiers from unannotated texts. In Proceedings of CICLing 2005. [find](https://scholar.google.com/scholar?q=Janyce+Wiebe+and+Ellen+Riloff%2E+2005%2E+Creating+subjective+and+objective+sentence+classifiers+from+unannotated+texts%2E+In+Proceedings+of+CICLing+2005%2E)
20. Theresa Wilson, Janyce Wiebe, and Paul Hoffmann. 2005. [Recognizing contextual polarity: An exploration of features for phrase-level sentiment analysis](https://md.research.elamur.ai/free/en/papers/wilson-2009-recognizing-contextual-polarity-exploration-features/). Computational linguistics, 35(3):399–433, 2009.
21. Yusuke Yanamoto. 2011. Twitter4j: A java library for the twitter api, 2011. [find](https://scholar.google.com/scholar?q=Yusuke+Yanamoto%2E+2011%2E+Twitter4j%3A+A+java+library+for+the+twitter+api%2C+2011%2E)
22. Hong Yu and Vasileios Hatzivassiloglou 2003. [Towards answering opinion questions: Separating facts from opinions and identifying the polarity of opinion sentences](https://md.research.elamur.ai/free/en/papers/yu-2003-answering-opinion-questions-separating-facts/) In EMNLP-2003, pages 129–136. Association for Computational Linguistics.
23. Rabih Zbib, Erika Malchiodi, Jacob Devlin, David Stallard, Spyros Matsoukas, Richard Schwartz, John Makhoul, Omar F. Zaidan and Chris Callison-Burch. 2012. [Machine translation of arabic dialects](https://md.research.elamur.ai/free/en/papers/zbib-2012-machine-translation-arabic-dialects/). In Proceedings of NAACL.

*Updated 30 October 2026*