Domain-Adapted BERT-based Models for Nuanced Arabic Dialect Identification and Tweet Sentiment Analysis

Bayrak, Gıyaseddin; Issifu, Abdul Majeed

doi:10.18653/v1/2022.wanlp-1.43

Cited by 4 publications

(2 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Various models were combined and performance was enhanced using a combination of TF-IDF and n-grams. An ensembling of transformerbased models, predominantly using variations of MARBERT was employed for dialect detection as well as sentiment analysis, in the NADI 2022 shared tasks (Bayrak and Issifu, 2022), (Khered et al, 2022). (Oumar and Mrini, 2022) addressed the issue regarding an imbalance in the classes of the NADI dataset by using focal loss and employed various Arabic BERT-based models.…”

Section: Related Workmentioning

confidence: 99%

Mavericks at NADI 2023 Shared Task: Unravelling Regional Nuances through Dialect Identification using Transformer-based Approach

Deshpande,

Patwardhan,

Deshpande

et al. 2023

Proceedings of ArabicNLP 2023

View full text Add to dashboard Cite

In this paper, we present our approach for the "Nuanced Arabic Dialect Identification (NADI) Shared Task 2023". We highlight our methodology for subtask 1 which deals with countrylevel dialect identification. Recognizing dialects plays an instrumental role in enhancing the performance of various downstream NLP tasks such as speech recognition and translation. The task uses the Twitter dataset (TWT-2023) that encompasses 18 dialects for the multi-class classification problem. Numerous transformerbased models, pre-trained on Arabic language, are employed for identifying country-level dialects. We fine-tune these state-of-the-art models on the provided dataset. The ensembling method is leveraged to yield improved performance of the system. We achieved an F 1 -score of 76.65 (11th rank on the leaderboard) on the test dataset.

show abstract

Section: Related Workmentioning

confidence: 99%

Mavericks at NADI 2023 Shared Task: Unravelling Regional Nuances through Dialect Identification using Transformer-based Approach

Deshpande,

Patwardhan,

Deshpande

et al. 2023

Proceedings of ArabicNLP 2023

View full text Add to dashboard Cite

show abstract

“…(Abdel-Salam, 2022) used is an ensemble between fine-tuned BERT-based models and various approaches of parameter efficient tuning including p-tuning and prompt-tuning. (Bayrak and Issifu, 2022) used general pre-training as a first step followed by fine-tuning. AlKhamissi et al (2021) added an adapter layer on top of the MARBERT model.…”

Section: Introductionmentioning

confidence: 99%

rematchka at NADI 2023 shared task: Parameter Efficient tuning for Dialect Identification and Dialect Machine Translation

Abdel-Salam

2023

Proceedings of ArabicNLP 2023

View full text Add to dashboard Cite

Dialect identification systems play a significant role in various fields and applications as in speech and language technologies, facilitating language education, supporting sociolinguistic research, preserving linguistic diversity, and enhancing text-to-speech systems. In this paper, we provide our findings and results in the NADI 2023 shared task for country-level dialect identification and machine translation (MT) from dialect to MSA. The proposed models achieved an F1-score of 86.18 at the dialect identification task, securing second place in the first subtask. Whereas for the machine translation task, the submitted model achieved a BLEU score of 11.37 securing fourth and third place in the second and third subtasks. The proposed model utilizes parameter-efficient training methods which achieves better performance when compared to conventional fine-tuning during the experimentation phase.

show abstract

Topic Classification of Text-Based Lesson Questions in Turkish with BERTurk

Doğan,

Sayar,

Çetiner

2023

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Domain-Adapted BERT-based Models for Nuanced Arabic Dialect Identification and Tweet Sentiment Analysis

Cited by 4 publications

References 0 publications

Mavericks at NADI 2023 Shared Task: Unravelling Regional Nuances through Dialect Identification using Transformer-based Approach

Mavericks at NADI 2023 Shared Task: Unravelling Regional Nuances through Dialect Identification using Transformer-based Approach

rematchka at NADI 2023 shared task: Parameter Efficient tuning for Dialect Identification and Dialect Machine Translation

Topic Classification of Text-Based Lesson Questions in Turkish with BERTurk

Contact Info

Product

Resources

About