---
language:
- ar
library_name: transformers
pipeline_tag: text-classification
metrics:
- accuracy
widget:
- text: ' جماعة انصار الله الحوثية تقوم بعدل حشد جماهيري ضد العدوان الغاشم على اليمن'
- text:  'ارتفاع اسعار المواد الغذائية في الاسواق مما يؤدي إلى ازمة في الاقتصاد'
- text: 'فوز المنتخب اليمني برياضة كرة القدم فاز بكاس غرب اسياء للناشئين'
- text: 'يف سيؤثر الذكاء الاصطناعي التوليدي على الصناعات العالمية الكبرى ف'

datasets:
- IBB-University/Ghadeer_news
---

# Classification Of Arabic News Using Arabert 
-One of the famous models that use transform networks in Arabic language classification is “BERT” (Bidirectional Encoder Representations from Transformers). BERT is trained on a huge amount of diverse linguistic data, including Arabic, which allows it to better understand language relationships. 
Other BERT-based models have been developed to improve classification performance in Arabic, such as “AraBERT” and “ARA-BERT”. These models are trained on large data specific to the Arabic language, allowing them to achieve outstanding classification performance for the Arabic language. 
The use of transfer networks in Arabic classification is currently an active area of research and development, with researchers and engineers working to improve existing models and develop new techniques to meet the challenges of Arabic and improve classification accuracy in this context.
# Google Scholar has our Bibtex wrong (missing name), use this instead

@inproceedings{antoun2020arabert,
  title={AraBERT: Transformer-based Model for Arabic Language Understanding},
  author={Antoun, Wissam and Baly, Fady and Hajj, Hazem},
  booktitle={LREC 2020 Workshop Language Resources and Evaluation Conference 11--16 May 2020},
  pages={9}
}
# DATASET

|       |              |
| :-------------------:  | :-----------:|
|  Local                 |5000          |
| Sports                 |5000          |
| Policy                 | 5000         |
|Economy                 |5000          | 
|Cultural                |5000          |          
|Technology              |5000          |   


# LABEL_DATASET

|         |               |                
| :--------------------------:  | :-----------:|
|lable_0                      |  رياضية      |
|lable_ سياسية |                      1       |    
|lable_اقتصاد  |                      2      | 
|lable_تكنولوجيا |                       3    |      
|lable_محلية  |                       4       |            
|lable_ثقافية  |                       5      |  




# Training parameters

|                       |              |
| :-------------------: | :-----------:|
|  Training batch size  |     `8`      |
| Evaluation batch size |     `8`      |
|     Learning rate     |    `2e-5`    |
|   Max length target   |    `203`     |
|           Epoch       |       `1 `    | 
|                       |              |

# # Results

|                         |               |
| :---------------------: | :-----------: | 
|   raining Loss:     |    `0.21533072472327064     |  
| Classification Accuracy |   `0.1619285045662197`    | 
|   Validation Accuracy:   |   `0.9664634146341463`    |
|                         |               |