---
title: "SLSA: A Sentiment Lexicon for Standard Arabic (2015)"
description: "To build it, they linked the free morphological analyzer AraMorph to SentiWordNet. 21 references, cited by 12 papers here."
authors: ["Ramy Eskander", "Owen Rambow"]
year: 2015
venue: "Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing"
source: "https://aclanthology.org/D15-1304/"
doi: "10.18653/v1/d15-1304"
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/eskander-2015-slsa-sentiment-lexicon-standard-arabic/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/eskander-2015-slsa-sentiment-lexicon-standard-arabic/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2015](https://md.research.elamur.ai/free/en/papers/2015/page/2/) / SLSA: A Sentiment Lexicon for Standard Arabic

# SLSA: A Sentiment Lexicon for Standard Arabic

**Authors:** [Ramy Eskander](https://md.research.elamur.ai/free/en/authors/ramy-eskander/) and [Owen Rambow](https://md.research.elamur.ai/free/en/authors/owen-rambow/) · **Year:** 2015 · **Venue:** [Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing](https://md.research.elamur.ai/free/en/venues/emnlp/) ([EMNLP 2015](https://md.research.elamur.ai/free/en/venues/emnlp/2015/)) · **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/D15-1304/) · [PDF](https://aclanthology.org/D15-1304.pdf) · [DOI 10.18653/v1/d15-1304](https://doi.org/10.18653/v1/d15-1304) · [OpenAlex](https://openalex.org/W2251526503)

**Citations:** 60 citations (OpenAlex, 2026-10-14) · [12](https://md.research.elamur.ai/free/en/papers/eskander-2015-slsa-sentiment-lexicon-standard-arabic/cited-by/) in our library, of the 149,439 papers whose bibliographies we read

## Summary
Arabic sentiment lexicons available before this work had small coverage, noisy or inflected entries, weak use of part of speech, or relied on resources that are not free. Eskander and Rambow built SLSA, a free lemma-level lexicon for Standard Arabic. It contains about 35,000 (lemma, POS) entries, and each one carries positive, negative and objective scores plus an English gloss. To build it, they linked the free morphological analyzer AraMorph to SentiWordNet. First they cleaned AraMorph and normalized its glosses through lemmatization, stop-word removal and stripping of derivational suffixes. Then they matched the gloss terms to SentiWordNet synset terms under a POS agreement condition that loosens in stages. The method uses no machine learning and only a few heuristics. Human judges compared SLSA with ArSenL, the closest earlier lexicon. SLSA was more accurate (80.1% vs. 68.0%) and also improved a lexicon-feature SVM sentiment classifier. The result is a reusable, freely licensed resource that inflected forms can look up through their lemma.

## Key points
- Linking runs as a cascade: exact gloss term plus POS links 83.6% of AraMorph entries; allowing NOUN/ADJ swaps and VERB→ADJ adds 6.7%; ignoring POS entirely adds 0.6%, with scores averaged across POS.
- On 400 random (lemma, POS) pairs, SLSA lacks only 4 (1%), while ArSenL lacks 103 (26%).
- On the entries both lexicons cover, accuracy is 51.9% for the neutral baseline, 68.0% for ArSenL and 80.1% for SLSA. This is a 37.8% error reduction over ArSenL.
- Extrinsic test: an SVM (LIBSVM) uses three features, the averaged positive, negative and objective lexicon scores, on subjective sentences from the Abdul-Mageed et al. (2011) corpus built on Penn Arabic Treebank part 1 v3. Weighted F1 is 68.6% with SLSA vs. 65.1% with ArSenL.
- Error analysis:
  - About 93% of SLSA errors come from doubtful scores already present in SentiWordNet.
  - Most uncovered lemmas (about 75%) name Arabic or Islamic concepts that have no English equivalent.

## Interesting ideas
- Link two lexicons through a sequence of matching rules, each looser than the last, so precise matches win and looser rules only fill gaps.
- Before matching, normalize the keys on both sides: lemmatize glosses, remove parenthetical notes, and strip a hand-made list of derivational suffixes, but only when the result is a real entry.
- When one word has several senses with different scores under the same POS, average them into a single (word, POS) value instead of guessing a sense.
- Let humans judge numeric scores by first collapsing them into a few intensity bins, have two annotators work blind to the source, and settle disagreements by discussion.
- Sort coverage gaps and errors into causes, such as missing concepts in the target resource, bad source glosses, or noisy upstream scores.

## For Mova

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

- **I1** (world): Mapping our lemmas onto WordNet/FrameNet/VerbNet entries could follow the same tiered rules, and each link would record which tier produced it, which keeps the step explainable.
- **I2** (tokens): A checked list of derivation rules (e.g. -ness → adjective) could map out-of-lexicon lemmas onto known base lemmas when we look up verb classes or ontology entries.
- **I3** (world): This gives a sensible default value for a lemma when sense disambiguation is uncertain, and that default could be marked as averaged so later steps know it is soft.
- **I4** (learn): We could evaluate graded outputs such as confidence or attribute values the same way, turning fuzzy numbers into checkable classes for gold and silver comparisons.
- **I5** (learn): Tagging each failure with whether the fault is in Mova or in the compiled knowledge underneath would direct repair work to the right component.

*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

- [Columbia University](https://md.research.elamur.ai/free/en/universities/columbia-university/) — University, New York, United States · 241 works · 54,053 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: Ramy Eskander and Owen Rambow. 2015. SLSA: A Sentiment Lexicon for Standard Arabic. Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing.

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

```bibtex
@inproceedings{eskander2015slsa,
  title = {{SLSA: A Sentiment Lexicon for Standard Arabic}},
  author = {Ramy Eskander and Owen Rambow},
  booktitle = {Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing},
  year = {2015},
  doi = {10.18653/v1/d15-1304},
  url = {https://doi.org/10.18653/v1/d15-1304}
}
```

- [← previous Cross-Lingual Dependency Parsing with Universal Dependencies and Predicted PoS Labels](https://md.research.elamur.ai/free/en/papers/tiedemann-2015-cross-lingual-dependency-parsing-universal/)
- [next → Context-aware Frame-Semantic Role Labeling](https://md.research.elamur.ai/free/en/papers/roth-2015-context-aware-frame-semantic-role/)
- [↑ back to papers of 2015](https://md.research.elamur.ai/free/en/papers/2015/page/2/)

## References

21 references, 7 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. 2012. Toward building a large-scale Arabic Sentiment Lexicon. In The Sixth International Global Word-Net Conference, Matsue, Japan. [find](https://scholar.google.com/scholar?q=Muhammad+Abdul%2DMageed+and+Mona+Diab%2E+2012%2E+Toward+building+a+large%2Dscale+Arabic+Sentiment+Lexicon%2E+In+The+Sixth+International+Global+Word%2DNet+Conference%2C+Matsue%2C+Japan%2E)
2. 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 The Ninth international conference on Language Resources and Evaluation (LREC), Reykjavik, Iceland.
3. 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 The 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies, Oregon, USA.
4. Samah Alhazmi, William Black, and John McNaught. 2013. SentiWordNet in Relation to SentiWordNet 3.0. International Journal of Computational Linguistics, 4. [find](https://scholar.google.com/scholar?q=Samah+Alhazmi%2C+William+Black%2C+and+John+McNaught%2E+2013%2E+SentiWordNet+in+Relation+to+SentiWordNet+3%2E0%2E+International+Journal+of+Computational+Linguistics%2C+4%2E)
5. 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 The Seventh international conference on Language Resources and Evaluation (LREC), Malta.
6. 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/). In Arabic Natural Language Processing Workshop (WANLP), EMNLP, Doha, Qatar.
7. William Black, Sabri Elkateb, Horacio Rodriguez, Musa Alkhalifa, Piek Vossen, Adam Pease, and Christiane Fellbaum. 2006. Introducing the Arabic WordNet Project. In The Third International Word-Net Conference (GWC), Jeju Island, Korea. [find](https://scholar.google.com/scholar?q=William+Black%2C+Sabri+Elkateb%2C+Horacio+Rodriguez%2C+Musa+Alkhalifa%2C+Piek+Vossen%2C+Adam+Pease%2C+and+Christiane+Fellbaum%2E+2006%2E+Introducing+the+Arabic+WordNet+Project%2E+In+The+Third+International+Word%2DNet+Con)
8. Tim Buckwalter. 2004. Buckwalter Arabic Morphological Analyzer Version 2.0. Linguistic Data Consortium, Philadelphia. [↗](https://doi.org/10.35111/050q-5r95)
9. Chih-Chung Chang and Chih-Jen Lin. 2011. Libsvm: A library for support vector machines. ACM Transactions on Intelligent Systems and Technology (TIST), 2. [find](https://scholar.google.com/scholar?q=Chih%2DChung+Chang+and+Chih%2DJen+Lin%2E+2011%2E+Libsvm%3A+A+library+for+support+vector+machines%2E+ACM+Transactions+on+Intelligent+Systems+and+Technology+%28TIST%29%2C+2%2E)
10. Mona Diab, Mohamed Al-Badrashiny, Maryam Aminian, Mohammed Attia, Pradeep Dasigi, Heba Elfardy, Ramy Eskander, Nizar Habash, Abdelati Hawwari, and Wael Salloum. 2014. [Tharwa: A Large Scale Dialectal Arabic - Standard Arabic - English Lexicon](https://md.research.elamur.ai/free/en/papers/diab-2014-tharwa-large-scale-dialectal-arabic/). In The Ninth international conference on Language Resources and Evaluation (LREC), Reykjavik, Iceland.
11. Alaa El-Halees. 2011. Arabic Opinion Mining Using Combined Classification Approach. In The International Arab Conference on Information Technology (ACIT), Amman, Jordan. [find](https://scholar.google.com/scholar?q=Alaa+El%2DHalees%2E+2011%2E+Arabic+Opinion+Mining+Using+Combined+Classification+Approach%2E+In+The+International+Arab+Conference+on+Information+Technology+%28ACIT%29%2C+Amman%2C+Jordan%2E)
12. Mohamed Elarnaoty, Samir AbdelRahman, and Aly Fahmy. 2012. A Machine Learning Approach For Opinion Holder Extraction in Arabic Language. In Corr, Amman, Jordan. [↗](https://doi.org/10.5121/ijaia.2012.3205)
13. Mohamed Elhawary and Mohamed Elfeky. 2010. Mining Arabic Business Reviews. In IEEE International Conference on Data Mining Workshops, Sydney, Australia. [find](https://scholar.google.com/scholar?q=Mohamed+Elhawary+and+Mohamed+Elfeky%2E+2010%2E+Mining+Arabic+Business+Reviews%2E+In+IEEE+International+Conference+on+Data+Mining+Workshops%2C+Sydney%2C+Australia%2E)
14. David Graff, Mohamed Maamouri, Basma Bouziri, Sondos Krouna, Seth Kulick, and Tim Buckwalter. 2009. Standard Arabic Morphological Analyzer (SAMA) Version 3.1. Linguistic Data Consortium LDC2009E73. [find](https://scholar.google.com/scholar?q=David+Graff%2C+Mohamed+Maamouri%2C+Basma+Bouziri%2C+Sondos+Krouna%2C+Seth+Kulick%2C+and+Tim+Buckwalter%2E+2009%2E+Standard+Arabic+Morphological+Analyzer+%28SAMA%29+Version+3%2E1%2E+Linguistic+Data+Consortium+LDC2009E73%2E)
15. Mohamed Maamouri, Ann Bies, Tim Buckwalter, and Wigdan Mekki. 2004. The penn arabic treebank: Building a large-scale annotated arabic corpus. In NEMLAR Conference on Arabic Language Resources and Tools, Cairo, Egypt. [find](https://scholar.google.com/scholar?q=Mohamed+Maamouri%2C+Ann+Bies%2C+Tim+Buckwalter%2C+and+Wigdan+Mekki%2E+2004%2E+The+penn+arabic+treebank%3A+Building+a+large%2Dscale+annotated+arabic+corpus%2E+In+NEMLAR+Conference+on+Arabic+Language+Resources+and+Tool)
16. Fawaz H.H. Mahyoub, Muazzam A. Siddiqui, and Mohamed Y. Dahab. 2014. Building an Arabic Sentiment Lexicon Using Semi-supervised Learning. Journal of King Saud University - Computer and Information Sciences, 26. [↗](https://doi.org/10.1016/j.jksuci.2014.06.003)
17. Christopher D. Manning, Mihai Surdeanu, John Bauer, Jenny Finkel, Steven J. Bethard, and David Mc-Closky. 2014. [The Stanford CoreNLP Natural Language Processing Toolkit](https://md.research.elamur.ai/free/en/papers/manning-2014-stanford-corenlp-natural-language-processing/). In The 52nd Annual Meeting of the Association for Computational Linguistics: System Demonstrations, Sydney, Australia.
18. George A. Miller, Richard Beckwith, and Derek Gross Christiane Fellbaum, and Katherine Miller. 1990. Introduction to WordNet: An On-line Lexical Database. International Journal of Lexicography, 3. [↗](https://doi.org/10.1093/ijl/3.4.235)
19. Hanaa Mobarz, Mohsen Rashwan, and Ibrahim Abdel-Rahman. 2011. Generating lexical Resources for Opinion Mining in Arabic Language Automatically. In The Eleventh Conference on Language Engineering (ESOLE), Cairo-Egypt. [find](https://scholar.google.com/scholar?q=Hanaa+Mobarz%2C+Mohsen+Rashwan%2C+and+Ibrahim+Abdel%2DRahman%2E+2011%2E+Generating+lexical+Resources+for+Opinion+Mining+in+Arabic+Language+Automatically%2E+In+The+Eleventh+Conference+on+Language+Engineering+%28ESOL)
20. Mike Thelwall, Kevan Buckley, Georgios Paltoglou, and Di Cai. 2010. Sentiment Strength Detection in Short Informal Text. Journal of the American Society for Information Science and Technology, 61. [↗](https://doi.org/10.1002/asi.21416)
21. 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 The First conference on Human Language Technology and Empirical Methods in Natural Language Processing (EMNLP), Vancouver, Canada.

*Updated 1 November 2026*