---
title: "University of Tripoli at AraSentEval — paper by Nwesri et al. (2026)"
description: "The paper describes the University of Tripoli entry for Subtask 1 of the AraSentEval 2026 shared task at OSACT7. 29 references."
authors: ["Abdusalam F. Ahmad Nwesri", "Amani Bahlul Sharif", "Sarah Farag S. Hmeid"]
year: 2026
venue: "The 7th Workshop on Open-Source Arabic Corpora and Processing Tools (OSACT7) with 5 Shared Tasks"
source: "https://aclanthology.org/2026.osact-1.42/"
doi: "10.63317/3zunt9qnt2i4"
licence: "CC BY-NC 4.0 (ELRA, LREC proceedings)"
licence_url: "https://creativecommons.org/licenses/by-nc/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/nwesri-2026-university-tripoli-arasenteval-fine-tuning/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/nwesri-2026-university-tripoli-arasenteval-fine-tuning/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2026](https://md.research.elamur.ai/free/en/papers/2026/page/47/) / University of Tripoli at AraSentEval: Fine-Tuning MARBERTv2 and CAMELBERT for Multi-Dialect Arabic Sentiment Analysis

# University of Tripoli at AraSentEval: Fine-Tuning MARBERTv2 and CAMELBERT for Multi-Dialect Arabic Sentiment Analysis

**Authors:** [Abdusalam F. Ahmad Nwesri](https://md.research.elamur.ai/free/en/authors/abdusalam-f-ahmad-nwesri/), [Amani Bahlul Sharif](https://md.research.elamur.ai/free/en/authors/amani-bahlul-sharif/) and Sarah Farag S. Hmeid · **Year:** 2026 · **Venue:** [The 7th Workshop on Open-Source Arabic Corpora and Processing Tools (OSACT7) with 5 Shared Tasks](https://md.research.elamur.ai/free/en/venues/osact/) ([Workshop on Open-Source Arabic Corpora and Processing Tools with Shared Tasks 2026](https://md.research.elamur.ai/free/en/venues/osact/2026/)) · **Publisher:** [Association for Computational Linguistics](https://md.research.elamur.ai/free/en/publishers/association-for-computational-linguistics/) · **Licence:** [CC BY-NC 4.0 (ELRA, LREC proceedings)](https://creativecommons.org/licenses/by-nc/4.0/)

**Links:** [Publisher's page](https://aclanthology.org/2026.osact-1.42/) · [PDF](https://aclanthology.org/2026.osact-1.42.pdf) · [DOI 10.63317/3zunt9qnt2i4](https://doi.org/10.63317/3zunt9qnt2i4) · [OpenAlex](https://openalex.org/W7166594779)

**Citations:** 0 citations (OpenAlex, 2026-10-14)

## Summary
The paper describes the University of Tripoli entry for Subtask 1 of the AraSentEval 2026 shared task at OSACT7. The task is three-way sentiment classification (positive, negative, neutral) of Arabic text in Moroccan (Darija), Egyptian, Jordanian and Saudi dialects. The data is the Multi-Dialect-Sent (MDS-3) collection of 1,731 hotel reviews, which is roughly balanced across the four dialects. The authors fine-tuned two dialect-aware pre-trained encoders, MARBERTv2 and a CAMeLBERT sentiment checkpoint, with the Hugging Face Trainer on Google Colab. They tuned the hyperparameters and compared raw text against a normalized version.

Normalization removed diacritics, unified letter variants and collapsed repeated characters. It consistently hurt results, and CAMeLBERT was affected more than MARBERTv2. The authors take this as evidence that informal orthography carries useful sentiment signal.

On the hidden test set, MARBERTv2 reached a Macro F1 of 0.843, ranking 4th of 13 teams. The work shows that a carefully tuned single dialect-oriented model, with little preprocessing, is competitive for low-resource dialectal sentiment analysis.

## Key points
- **Data:** MDS-3 contains 1,731 labelled training reviews and a 312-review test set. The organizers' official metric is Macro F1.
- **Models:** MARBERTv2 and `CAMeL-Lab/bert-base-arabic-camelbert-mix-sentiment`.
- **Training setup:** 80/20 train/validation split, max length 128, learning rate 1e-4, batch size 16, weight decay 1e-4, warmup 0.1, and 8 epochs.
- **Validation F1 on raw text:** MARBERTv2 scored 0.965 and CAMeLBERT 0.960. Both models scored lower on preprocessed text.
- **Test results:** MARBERTv2 scored 0.843 and CAMeLBERT 0.824. The large drop from validation scores points to limited generalization from the small dataset.

*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

- [University of Tripoli](https://md.research.elamur.ai/free/en/universities/university-of-tripoli/) — University, Tripoli, Libya · 5 works · 8 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 · [🎭 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

🧭 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: Abdusalam F. Ahmad Nwesri, Amani Bahlul Sharif and Sarah Farag S. Hmeid. 2026. University of Tripoli at AraSentEval: Fine-Tuning MARBERTv2 and CAMELBERT for Multi-Dialect Arabic Sentiment Analysis. The 7th Workshop on Open-Source Arabic Corpora and Processing Tools (OSACT7) with 5 Shared Tasks.

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

```bibtex
@inproceedings{nwesri2026university,
  title = {{University of Tripoli at AraSentEval: Fine-Tuning MARBERTv2 and CAMELBERT for Multi-Dialect Arabic Sentiment Analysis}},
  author = {Abdusalam F. Ahmad Nwesri, Amani Bahlul Sharif and Sarah Farag S. Hmeid},
  booktitle = {The 7th Workshop on Open-Source Arabic Corpora and Processing Tools (OSACT7) with 5 Shared Tasks},
  year = {2026},
  doi = {10.63317/3zunt9qnt2i4},
  url = {https://doi.org/10.63317/3zunt9qnt2i4}
}
```

- [← previous Universally Empowering Zeroth-Order Optimization via Adaptive Layer-wise Sampling](https://md.research.elamur.ai/free/en/papers/wang-2026-universally-empowering-zeroth-order-optimization/)
- [next → University Speaking for Everyone: Assessing Changes in Italian Higher Education Statutes toward Gender-Inclusive Language](https://md.research.elamur.ai/free/en/papers/salto-2026-university-speaking-everyone-assessing-changes/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/47/)

## References

29 references, 7 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. 2010. Encyclopedia of machine learning. [find](https://scholar.google.com/scholar?q=2010%2E+Encyclopedia+of+machine+learning%2E)
2. Mohamed Elhag Mohamed Abo, Ram Gopal Raj, and Atika Qazi. 2019. A review on arabic sentiment analysis: State-of-the-art, taxonomy and open research challenges. IEEE Access, 7:162008–162024. [↗](https://doi.org/10.1109/access.2019.2951530)
3. Hend Al-Khalifa. 2025. [iwan-nlp at ahasis 2025: A stacked ensemble of arabic transformers for sentiment analysis on arabic dialects in the hospitality domain](https://md.research.elamur.ai/free/en/papers/khalifa-2025-iwan-nlp-ahasis-2025-stacked/). In Proceedings of the Shared Task on Sentiment Analysis for Arabic Dialects, pages 7–13, Varna, Bulgaria. INCOMA Ltd., Shoumen, Bulgaria.
4. Tareq Al-Moslmi, M. Albared, Adel Al-Shabi, N. Omar, and S. Abdullah. 2018. Arabic senti-Preprocessed Dataset Acc. F1 P R 0.960 0.958 0.953 0.955 0.936 0.932 0.934 0.934 lexicon: Constructing publicly available language resources for arabic sentiment analysis. Journal of Information Science, 44:345 – 362. [find](https://scholar.google.com/scholar?q=Tareq+Al%2DMoslmi%2C+M%2E+Albared%2C+Adel+Al%2DShabi%2C+N%2E+Omar%2C+and+S%2E+Abdullah%2E+2018%2E+Arabic+senti%2DPreprocessed+Dataset+Acc%2E+F1+P+R+0%2E960+0%2E958+0%2E953+0%2E955+0%2E936+0%2E932+0%2E934+0%2E934+lexicon%3A+Constructing+publicly)
5. Amani A. Aladeemy, Ali Alzahrani, Mohammad H. Algarni, S. N. Alsubari, Theyazn H. H. Aldhyani, Sachin N. Deshmukh, O. I. Khalaf, Wing-Keung Wong, and Sameer Aqburi. 2024. Advancements and challenges in arabic sentiment analysis: A decade of methodologies, applications, and resource development. Heliyon, 10. [↗](https://doi.org/10.1016/j.heliyon.2024.e39786)
6. D. Alahmadi. 2025. Human versus ai: A comparative study of zero-shot llms and transformer models against human annotations for arabic sentiment analysis. International Journal of Advanced Computer Science and Applications. [↗](https://doi.org/10.14569/ijacsa.2025.0160882)
7. Alhanouf Alduailej and Abdulrahman Alothaim. 2022. Araxlnet: pre-trained language model for sentiment analysis of arabic. Journal of Big Data, 9. [↗](https://doi.org/10.1186/s40537-022-00625-z)
8. Maram Alharbi, Saad Ezzini, Hansi Hettiarachchi, Tharindu Ranasinghe, and Ruslan Mitkov. 2025a. Evaluating large language models on arabic dialect sentiment analysis. In Proceedings of the 15th International Conference on Recent Advances in Natural Language Processing (RANLP). [find](https://scholar.google.com/scholar?q=Maram+Alharbi%2C+Saad+Ezzini%2C+Hansi+Hettiarachchi%2C+Tharindu+Ranasinghe%2C+and+Ruslan+Mitkov%2E+2025a%2E+Evaluating+large+language+models+on+arabic+dialect+sentiment+analysis%2E+In+Proceedings+of+the+15th+Intern)
9. Maram I. Alharbi, Salmane Chafik, Saad Ezzini, Ruslan Mitkov, Tharindu Ranasinghe, and Hansi Hettiarachchi. 2025b. [AHaSIS: Shared task on sentiment analysis for Arabic dialects](https://md.research.elamur.ai/free/en/papers/alharbi-2025-ahasis-shared-task-sentiment-analysis/). In Proceedings of the Shared Task on Sentiment Analysis for Arabic Dialects, pages 1–6, Varna, Bulgaria. INCOMA Ltd., Shoumen, Bulgaria.
10. Maram I Alharbi, Saad Ezzini, Hansi Hettiarachchi, Tharindu Ranasinghe, and Ruslan Mitkov. 2025c. [Ador: Dataset for arabic dialects in hotel reviews: A human benchmark for sentiment analysis](https://md.research.elamur.ai/free/en/papers/alharbi-2025-ador-dataset-arabic-dialects-hotel/). In Proceedings of the First Workshop on Advancing NLP for Low-Resource Languages, pages 187– 191.
11. Yusuf Aliyu, A. Sarlan, Kamaluddeen Usman Danyaro, A. S. B. A. Rahman, and Mujaheed Abdullahi. 2024. Sentiment analysis in low-resource settings: A comprehensive review of approaches, languages, and data sources. IEEE Access, 12:66883–66909. [↗](https://doi.org/10.1109/access.2024.3398635)
12. Alanoud Alotaibi and Farrukh Nadeem. 2025. An unsupervised integrated framework for arabic aspect-based sentiment analysis and abstractive text summarization of traffic services using transformer models. Smart Cities. [↗](https://doi.org/10.3390/smartcities8020062)
13. Ohud Alsemaree, A. Alam, S. Gill, and Steve Uhlig. 2024. An analysis of customer perception using lexicon-based sentiment analysis of arabic texts framework. Heliyon, 10. [↗](https://doi.org/10.1016/j.heliyon.2024.e30320)
14. Adel M. A. Assiri, Ahmed Z. Emam, and Hmood Al-Dossari. 2018. Towards enhancement of a lexicon-based approach for saudi dialect sentiment analysis. Journal of Information Science, 44:184 – 202. [↗](https://doi.org/10.1177/0165551516688143)
15. 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 Proceedings of the EMNLP 2014 Workshop on Arabic Natural Language Processing (ANLP), pages 165–173. Association for Computational Linguistics.
16. Ekaba Bisong. 2019. Building Machine Learning and Deep Learning Models on Google Cloud Platform. Apress, Berkeley, CA. [↗](https://doi.org/10.1007/978-1-4842-4470-8)
17. Ghizlane Bourahouat, Manar Abourezq, and N. Daoudi. 2024. Improvement of moroccan dialect sentiment analysis using arabic bert-based models. Journal of Computer Science. [↗](https://doi.org/10.3844/jcssp.2024.157.167)
18. Saad Ezzini, Shadi Abudalfa, Maram Alharbi, Salmane Chafik, Samaher Alghamdi, Reem Alotaibi, Hamzah Luqman, Mo El-Haj, and Paul Rayson. 2026. AraSentEval: A shared task on sentiment analysis and swapping in arabic. In Proceedings of the 7th Workshop on Open-Source Arabic Corpora and Processing Tools (OSACT7), co-located with the Language Resources and Evaluation Conference (LREC 2026), Palma, Mallorca, Spain. [find](https://scholar.google.com/scholar?q=Saad+Ezzini%2C+Shadi+Abudalfa%2C+Maram+Alharbi%2C+Salmane+Chafik%2C+Samaher+Alghamdi%2C+Reem+Alotaibi%2C+Hamzah+Luqman%2C+Mo+El%2DHaj%2C+and+Paul+Rayson%2E+2026%2E+AraSentEval%3A+A+shared+task+on+sentiment+analysis+and+swapp)
19. Mustapha Jaballah, Dhaou Ghoul, and Ammar Mars. 2025. [Fine-tuning arabert model for arabic sentiment detection](https://md.research.elamur.ai/free/en/papers/jaballah-2025-fine-tuning-arabert-model-arabic/). In Proceedings of the Shared Task on Sentiment Analysis for Arabic Dialects, pages 14–23, Varna, Bulgaria. INCOMA Ltd., Shoumen, Bulgaria.
20. Areej Jaber, Israa Bahati, and Paloma Martínez. 2025. Leveraging pre-trained embeddings in an ensemble machine learning approach for arabic sentiment analysis. Frontiers in Artificial Intelligence, 8. [↗](https://doi.org/10.3389/frai.2025.1653728)
21. Souha Al Katat, Chamseddine Zaki, Hussein Hazimeh, Ibrahim El Bitar, Rafael Angarita, and Lionel Trojman. 2024. Natural language processing for arabic sentiment analysis: A systematic literature review. IEEE Transactions on Big Data, 10:576–594. [↗](https://doi.org/10.1109/tbdata.2024.3366083)
22. Omar Mansour, Eman Aboelela, Remon Talaat, and Mahmoud Bustami. 2025. Transformer-based ensemble model for dialectal arabic sentiment classification. PeerJ Computer Science, 11. [↗](https://doi.org/10.7717/peerj-cs.2644)
23. Yassir Matrane, F. Benabbou, and N. Sael. 2023. A systematic literature review of arabic dialect sentiment analysis. J. King Saud Univ. Comput. Inf. Sci., 35:101570. [↗](https://doi.org/10.1016/j.jksuci.2023.101570)
24. Abdullah Y. Muaad, Md Belal Bin Heyat, F. Akhtar, Usman Naseem, Wadeea R Naji, Suresha Mallappa, and H. J. 2025. Artificial intelligence for text analysis in the arabic and related middle eastern languages: Progress, trends, and future recommendations. Int. J. Intell. [↗](https://doi.org/10.1155/int/6091900)
25. A. Mutawa and S. Sruthi. 2025. A comparative evaluation of transformers and deep learning models for arabic meter classification. Applied Sciences. [↗](https://doi.org/10.3390/app15094941)
26. Abdusalam F. Ahmad Nwesri, Nabila Almabrouk S. Shinbir, and Amani Bahlul Sharif. 2025. [Sentiment analysis on arabic dialects: A multi-dialect benchmark](https://md.research.elamur.ai/free/en/papers/nwesri-2025-sentiment-analysis-arabic-dialects-multi/). In Proceedings of the Shared Task on Sentiment Analysis for Arabic Dialects, pages 86–91, Varna, Bulgaria. INCOMA Ltd., Shoumen, Bulgaria.
27. Oumaima Oueslati, E. Cambria, M. Hajhmida, and Habib Ounelli. 2020. [A review of sentiment analysis research in arabic language](https://md.research.elamur.ai/free/en/papers/oueslati-2020-review-sentiment-analysis-research-arabic/). ArXiv, abs/2005.12240. [↗](https://arxiv.org/abs/2005.12240)
28. Raheel Qader, François Portet, and Cyril Labbé. 2019. Semi-supervised neural text generation by joint learning of natural language generation and natural language understanding models. In Proceedings of the 12th International Conference on Natural Language Generation, pages 552– 562. [↗](https://doi.org/10.48550/arxiv.1910.03484)
29. Arwa A. Al Shamsi and Sherief Abdallah. 2021. Text mining techniques for sentiment analysis of arabic dialects: Literature review. Advances in Science, Technology and Engineering Systems Journal, 6:1012–1023. [↗](https://doi.org/10.25046/aj0601112)

*Updated 1 November 2026*