---
license: mit
datasets:
- Abdou/dz-sentiment-yt-comments
language:
- ar
metrics:
- f1
- accuracy
library_name: transformers
---
# BERT Models Fine-tuned on Algerian Dialect Sentiment Analysis
These are different BERT models (BERT Arabic models are initialized from [AraBERT](https://huggingface.co/aubmindlab/bert-large-arabertv02)) fine-tuned on the [Algerian Dialect Sentiment Analysis](https://huggingface.co/datasets/Abdou/dz-sentiment-yt-comments) dataset. The dataset contains 50,016 comments from YouTube videos in Algerian dialect. The models are evaluated on the testing set:


| Model Version       | No. of Parameters | Training Time  | F1-Score | Accuracy |
| ------------------- | ----------------- | -------------- | -------- | -------- |
| LSTM                | ~4 M              | 3 min          | 0.7399   | 0.7445   |
| Bi-LSTM             | ~4.3 M            | 6 min 35 s     | 0.7380   | 0.7437   |
| [BERT Base](https://huggingface.co/bert-base-uncased)           | ~109.5 M          | 33 min 20 s    | 0.6979   | 0.7500   |
| [BERT Large](https://huggingface.co/bert-large-uncased)          | ~335.1 M          | 1 h 50 min     | 0.6976   | 0.7484   |
| [BERT Arabic Mini](https://huggingface.co/Abdou/arabert-mini-algerian)    | ~11.6 M           | 2 min 40 s     | 0.7057   | 0.7527   |
| [BERT Arabic Medium](https://huggingface.co/Abdou/arabert-medium-algerian)  | ~42.1 M           | 11 min 25 s    | 0.7521   | 0.7860   |
| [BERT Arabic Base](https://huggingface.co/Abdou/arabert-base-algerian)    | ~110.6 M          | 34 min 19 s    | 0.7688   | 0.8002   |
| **[BERT Arabic Large](https://huggingface.co/Abdou/arabert-large-algerian)**   | **~336.7 M**          | **1 h 53 min**     | **0.7838**   | **0.8174**   |

# Citation
If you find our work useful, please cite it as follows:
```bibtex
@article{2023,
  title={Sentiment Analysis on Algerian Dialect with Transformers},
  author={Zakaria Benmounah and Abdennour Boulesnane and Abdeladim Fadheli and Mustapha Khial},
  journal={Applied Sciences},
  volume={13},
  number={20},
  pages={11157},
  year={2023},
  month={Oct},
  publisher={MDPI AG},
  DOI={10.3390/app132011157},
  ISSN={2076-3417},
  url={http://dx.doi.org/10.3390/app132011157}
}
```