Elamur Research⌕
☰
home / papers / 2017 / Modern Trends in Arabic Sentiment Analysis: A Survey

Modern Trends in Arabic Sentiment Analysis: A Survey

Paper · 2017 · Hala Mulki, Hatem Haddad and Ismail Babaoğlu

Summary

After the 2010 uprisings, Arabic opinion content on social media grew sharply, and demand for automatic opinion mining grew with it. Yet Arabic Sentiment Analysis (ASA) systems had rarely been compared directly. This survey, published in TAL (vol. 58, no. 3, 2017), reviews ASA work from roughly the preceding five years.

It first describes the standard sentiment pipeline:

  • noise cleaning and morphological preprocessing;
  • feature building, either hand-crafted n-gram and lexicon scores or learned text embeddings;
  • subjectivity detection followed by polarity classification, at word, aspect, sentence or document granularity.

It then lists what makes Arabic difficult: root-and-pattern morphology, few public datasets and lexicons, negation and sarcasm, Arabizi written with Latin letters and digits, and large differences between dialects. The reviewed systems are grouped by approach (classical machine learning, lexicon-based, deep learning) and by analysis level. Their reported results are compared to show which combination of method, resource and dataset tends to perform best. The paper's main value is as a map of techniques and reusable Arabic resources for future work.

Key points

  • Arabic is treated as three variants: Classical Arabic, Modern Standard Arabic (MSA) and Dialectal Arabic (DA). DA makes up most opinionated online text.
  • The survey separates "hand-crafted" features (binary or frequency-weighted bag-of-words and n-grams, lexicon intensity scores) from embedding features used in deep models.
  • Morphological analysis and disambiguation tools such as MADAMIRA (and the analyzer SAMA) produce, in one pass, the single most likely segmentation together with part-of-speech labels, stems, lemmas and restored diacritics.
  • The dialect examples are striking. One Levantine blessing ("May God grant you health") reads as a curse ("Burn in hell") in Tunisian. A Levantine negation particle can also act as an interrogative tag.
  • The bibliography covers resources and benchmarks such as ASTD, OCA, the Qatar Arabic Language Bank guidelines and SemEval-2017 Task 4 (Arabic track).

Interesting ideas

  • I1. Deciding whether a unit is subjective at all, and only then assigning polarity, keeps neutral factual text out of the polarity step.
  • I2. One particle can be a negator or a question tag, so negation must be resolved from its syntactic role rather than looked up in a word list.
  • I3. Sarcasm appears as a positive word ("lucky me") placed after a clearly bad situation (two hours of waiting, tickets sold out). The cue is the clash between the stated evaluation and the described outcome.
  • I4. The same expression can carry opposite sentiment in different dialects, so sentiment lexicons and stop/negation lists should be tied to a variety rather than shared globally.

For Mova

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

  • I1 (frames): add a "carries an attitude?" gate before attaching evaluation or polarity to a speaker's frame, so plain reports are not scored as opinions.
  • I2 (frames): handle English tag questions ("isn't it?") as a speech-act marker that does not flip the proposition's polarity, using the UD attachment to tell them apart.
  • I3 (frames): flag irony when an evaluative word's polarity contradicts the outcome state recorded for the same event in the text's world.
  • I4 (world): store evaluation values in the compiled knowledge with a register, dialect or domain attribute, so a value is applied only in the context where it holds.

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, topics and concepts

📚 Shelves 4The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksArtificial Intelligence · 16,194 worksArts and Humanities · 16,039 worksLanguage and Linguistics · 15,891 works

🧭 Topics 2Research subjects and languages the work is about.Sentiment & emotion · 972 worksSemitic languages · 2,908 works

Cite

Cite as Hala Mulki, Hatem Haddad and Ismail Babaoğlu. 2017. Modern Trends in Arabic Sentiment Analysis: A Survey. Dépôt institutionnel de l'Université libre de Bruxelles (Université Libre de Bruxelles).

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

@inproceedings{mulki2017modern,
  title = {{Modern Trends in Arabic Sentiment Analysis: A Survey}},
  author = {Hala Mulki, Hatem Haddad and Ismail Babaoğlu},
  booktitle = {Dépôt institutionnel de l'Université libre de Bruxelles (Université Libre de Bruxelles)},
  year = {2017},
  url = {https://aclanthology.org/2017.tal-3.3/}
}

References

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

  1. Abdul-Mageed M., Subjectivity and sentiment analysis of Arabic as a morophologically-rich language, PhD thesis, Indiana University, 2015. find
  2. Abdul-Mageed M., Diab M., Korayem M., “Subjectivity and sentiment analysis of modern standard Arabic”, Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies: short papers-Volume 2, Association for Computational Linguistics, p. 587-591, 2011.
  3. Abdul-Mageed M., Diab M., Kübler S., “SAMAR: Subjectivity and sentiment analysis for Arabic social media”, Computer Speech & Language, vol. 28, no 1, p. 20-37, 2014. find
  4. Abdulla N. A., Ahmed N. A., Shehab M. A., Al-Ayyoub M., “Arabic sentiment analysis: Lexicon-based and corpus-based”, 2013 IEEE Jordan Conference on Applied Electrical Engineering and Computing Technologies (AEECT), IEEE, p. 1-6, 2013. find
  5. Abdulla N., Mohammed S., Al-Ayyoub M., Al-Kabi M. et al., “Automatic lexicon construction for arabic sentiment analysis”, 2014 International Conference on Future Internet of Things and Cloud (FiCloud), IEEE, p. 547-552, 2014. find
  6. Al-Kabi M. N., Abdulla N. A., Al-Ayyoub M., “An analytical study of arabic sentiments: Maktoob case study”, 8th International Conference for Internet Technology and Secured Transactions (ICITST), IEEE, p. 89-94, 2013. find
  7. Al-Moslmi T., Albared M., Al-Shabi A., Omar N., Abdullah S., “Arabic senti-lexicon: Constructing publicly available language resources for Arabic sentiment analysis”, Journal of Information Science, 2017. ↗
  8. Arabic Sentiment Analysis: A Survey Al-Osaimi S., Badruddin K. M., “Role of Emotion icons in Sentiment classification of Arabic Tweets”, Proceedings of the 6th international conference on management of emergent digital ecosystems, ACM, p. 167-171, 2014. find
  9. Al-Rfou R., Perozzi B., Skiena S., “Polyglot: Distributed word representations for multilingual nlp”, arXiv preprint arXiv:1307.1662, 2013. ↗
  10. Al Sallab A. A., Baly R., Badaro G., Hajj H., El Hajj W., Shaban K. B., “Deep learning models for sentiment analysis in arabic”, ANLP Workshop, vol. 9, 2015. ↗
  11. Al Sallab A. A., Baly R., Hajj H., Shaban K. B., El-Hajj W., Badaro G., “AROMA: A Recursive Deep Learning Model for Opinion Mining in Arabic As a Low Resource Language”, ACM Trans. Asian Low-Resour. Lang. Inf. Process., vol. 16, no 4, p. 25:1-25:20, 2017. find
  12. Al-Twairesh N., Al-Khalifa H. S., Alsalman A., “AraSenTi: Large-Scale Twitter-Specific Arabic Sentiment Lexicons”, ACL (1), 2016.
  13. Altowayan A. A., Tao L., “Word embeddings for Arabic sentiment analysis”, 2016 IEEE International Conference on Big Data (Big Data), IEEE, p. 3820-3825, 2016. find
  14. Aly M. A., Atiya A. F., “LABR: A Large Scale Arabic Book Reviews Dataset.”, ACL (2), p. 494-498, 2013. ↗
  15. Assiri A., Emam A., Al-Dossari H., “Towards enhancement of a lexicon-based approach for Saudi dialect sentiment analysis”, Journal of Information Science, 2017. ↗
  16. Assiri A., Emam A., Aldossari H., “Arabic sentiment analysis: a survey”, International Journal of Advanced Computer Science and Applications, vol. 6, no 12, p. 75-85, 2015. find
  17. Badaro G., Baly R., Hajj H., Habash N., El-Hajj W., “A large scale Arabic sentiment lexicon for Arabic opinion mining”, ANLP 2014, 2014.
  18. Balikas G., Amini M.-R., “TwiSE at SemEval-2016 Task 4: Twitter Sentiment Classification”, arXiv preprint arXiv:1606.04351, 2016. ↗
  19. Baly R., Badaro G., Hamdi A., Moukalled R., Aoun R., El-Khoury G., Al Sallab A. A., Hajj H., Habash N., Shaban K., El-Hajj W., “OMAM at SemEval-2017 Task 4: Evaluation of English State-of-the-Art Sentiment Analysis Models for Arabic and a New Topic-based Model”, Proceedings of the 11th International Workshop on Semantic Evaluation (SemEval-2017), Association for Computational Linguistics, Vancouver, Canada, p. 603-610, August, 2017. ↗
  20. Banea C., Mihalcea R., Wiebe J., “Multilingual subjectivity: Are more languages better?”, Proceedings of the 23rd international conference on computational linguistics, Association for Computational Linguistics, p. 28-36, 2010. find
  21. Baniata L. H., Park S.-B., “Sentence Representation Network for Arabic Sentiment Analysis”, Proceedings of the Korean Information Science Society, 2016. find
  22. Biltawi M., Etaiwi W., Tedmori S., Hudaib A., Awajan A., “Sentiment classification techniques for Arabic language: A survey”, 7th International Conference on Information and Communication Systems (ICICS), IEEE, p. 339-346, 2016. find
  23. Collobert R., Weston J., Bottou L., Karlen M., Kavukcuoglu K., Kuksa P., “Natural language processing (almost) from scratch”, Journal of Machine Learning Research, vol. 12, p. 2493-2537, 2011.
  24. Diab M., “Second generation AMIRA tools for Arabic processing: Fast and robust tokenization, POS tagging, and base phrase chunking”, 2nd International Conference on Arabic Language Resources and Tools, vol. 110, 2009. find
  25. Duwairi R., Ahmed N. A., Al-Rifai S. Y., “Detecting sentiment embedded in Arabic social media–a lexicon-based approach”, Journal of Intelligent & Fuzzy Systems, vol. 29, no 1, p. 107-117, 2015. ↗
  26. Duwairi R., Marji R., Sha’ban N., Rushaidat S., “Sentiment analysis in arabic tweets”, 5th international conference on Information and communication systems (icics), IEEE, p. 1-6, 2014. find
  27. El-Beltagy S. R., Ali A., “Open issues in the sentiment analysis of Arabic social media: A case study”, 9th international conference onInnovations in information technology (iit), IEEE, p. 215-220, 2013. find
  28. El-Beltagy S. R., El kalamawy M., Soliman A. B., “NileTMRG at SemEval-2017 Task 4: Arabic Sentiment Analysis”, Proceedings of the 11th International Workshop on Semantic Evaluation (SemEval-2017), Association for Computational Linguistics, Vancouver, Canada, p. 790-795, August, 2017.
  29. Farra N., McKeown K., “SMARTies: Sentiment Models for Arabic Target entities”, arXiv preprint arXiv:1701.03434, 2017. ↗
  30. Farra N., McKeown K., Habash N., “Annotating targets of opinions in arabic using crowdsourcing”, Proceedings of the Second Workshop on Arabic Natural Language Processing, p. 89-98, 2015. ↗
  31. Gomez L., Patel Y., Rusiñol M., Karatzas D., Jawahar C., “Self-supervised learning of visual features through embedding images into text topic spaces”, arXiv preprint arXiv:1705.08631, 2017. ↗
  32. Gridach M., Haddad H., Mulki H., “Empirical Evaluation of Word Representations on Arabic Sentiment Analysis”, International Conference on Arabic Language Processing, Springer, p. 147-158, 2017. find
  33. Habash N. Y., “Introduction to Arabic natural language processing”, Synthesis Lectures on Human Language Technologies, vol. 3, no 1, p. 1-187, 2010. ↗
  34. Joulin A., Grave E., Bojanowski P., Mikolov T., “Bag of tricks for efficient text classification”, arXiv preprint arXiv:1607.01759, 2016. ↗
  35. Kiritchenko S., Zhu X., Mohammad S. M., “Sentiment analysis of short informal texts”, Journal of Artificial Intelligence Research, vol. 50, p. 723-762, 2014. ↗
  36. Le Q., Mikolov T., “Distributed representations of sentences and documents”, Proceedings of the 31st International Conference on Machine Learning (ICML-14), p. 1188-1196, 2014. ↗
  37. Liu B., “Sentiment analysis and opinion mining”, Synthesis lectures on human language technologies, vol. 5, no 1, p. 1-167, 2012. ↗
  38. Maamouri M., Graff D., Bouziri B., Krouna S., Bies A., Kulick S., “Standard Arabic morphological analyzer (SAMA) version 3.1”, Linguistic Data Consortium, Catalog No.: LDC2010L01, 2010. find
  39. Mdhaffar S., Bougares F., Esteve Y., Hadrich-Belguith L., “Sentiment Analysis of Tunisian Dialect: Linguistic Resources and Experiments”, WANLP 2017 (co-located with EACL 2017), 2017. find
  40. Mikolov T., Sutskever I., Chen K., Corrado G. S., Dean J., “Distributed representations of words and phrases and their compositionality”, Advances in neural information processing systems, p. 3111-3119, 2013. ↗
  41. Arabic Sentiment Analysis: A Survey Mohammad S. M., Kiritchenko S., Zhu X., “NRC-Canada: Building the state-of-the-art in sentiment analysis of tweets”, arXiv preprint arXiv:1308.6242, 2013. ↗
  42. Mohammad S. M., Salameh M., Kiritchenko S., “How Translation Alters Sentiment.”, J. Artif. Intell. Res. (JAIR), vol. 55, p. 95-130, 2016. ↗
  43. Mourad A., Darwish K., “Subjectivity and Sentiment Analysis of Modern Standard Arabic and Arabic Microblogs”, WASSA@ NAACL-HLT, p. 55-64, 2013.
  44. Nabil M., Aly M. A., Atiya A. F., “ASTD: Arabic Sentiment Tweets Dataset”, EMNLP, p. 2515-2519, 2015. ↗
  45. Pasha A., Al-Badrashiny M., Diab M. T., El Kholy A., Eskander R., Habash N., Pooleery M., Rambow O., Roth R., “MADAMIRA: A Fast, Comprehensive Tool for Morphological Analysis and Disambiguation of Arabic”, LREC, vol. 14, p. 1094-1101, 2014.
  46. Pennington J., Socher R., Manning C. D., “Glove: Global vectors for word representation”, EMNLP, vol. 14, p. 1532-1543, 2014.
  47. Piryani R., Madhavi D., Singh V. K., “Analytical mapping of opinion mining and sentiment analysis research during 2000–2015”, Information Processing & Management, vol. 53, no 1, p. 122-150, 2017. ↗
  48. Popescu A.-M., Etzioni O., “Extracting product features and opinions from reviews”, Natural language processing and text mining, Springer, p. 9-28, 2007. ↗
  49. Refaee E., Rieser V., “An Arabic Twitter Corpus for Subjectivity and Sentiment Analysis”, LREC, p. 2268-2273, 2014.
  50. Reyes A., Rosso P., “On the difficulty of automatically detecting irony: beyond a simple case of negation”, Knowledge and Information Systems, vol. 40, no 3, p. 595-614, 2014. ↗
  51. Rosenthal S., Farra N., Nakov P., “SemEval-2017 Task 4: Sentiment Analysis in Twitter”, Proceedings of the 11th International Workshop on Semantic Evaluation, SemEval ’17, Association for Computational Linguistics, Vancouver, Canada, August, 2017.
  52. Rushdi-Saleh M., Martín-Valdivia M. T., Ureña-López L. A., Perea-Ortega J. M., “OCA: Opinion corpus for Arabic”, Journal of the Association for Information Science and Technology, vol. 62, no 10, p. 2045-2054, 2011. ↗
  53. Salamah J. B., Elkhlifi A., “Microblogging opinion mining approach for kuwaiti dialect”, Computing Technology and Information Management, vol. 1, no 1, p. 9, 2016. find
  54. Salameh M., Mohammad S., Kiritchenko S., “Sentiment after Translation: A Case-Study on Arabic Social Media Posts”, HLT-NAACL, p. 767-777, 2015.
  55. Shoukry A., Rafea A., “Sentence-level Arabic sentiment analysis”, 2012 International Conference on Collaboration Technologies and Systems (CTS), IEEE, p. 546-550, 2012. find
  56. Turney P. D., “Thumbs up or thumbs down?: semantic orientation applied to unsupervised classification of reviews”, Proceedings of the 40th annual meeting on association for computational linguistics, Association for Computational Linguistics, p. 417-424, 2002. — Thumbs Up or Thumbs Down? Semantic Orientation Applied to Unsupervised Classification of Reviews ↗
  57. Zaghouani W., Habash N., Mohit B., The Qatar Arabic language bank guidelines, Technical report, Technical Report CMU-CS-QTR-124, School of Computer Science, Carnegie Mellon University, Pittsburgh, PA, September, 2014. find
  58. Zahran M. A., Magooda A., Mahgoub A. Y., Raafat H. M., Rashwan M., Atyia A., “Word Representations in Vector Space and their Applications for Arabic”, CICLing (1), p. 430-443, 2015. find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/mulki-2017-modern-trends-arabic-sentiment-analysis/