Omartificial-Intelligence-Space's picture
Update README.md
61d85c7 verified
metadata
dataset_info:
  features:
    - name: query
      dtype: string
    - name: answer
      dtype: string
  splits:
    - name: train
      num_bytes: 91841599
      num_examples: 100231
  download_size: 47254120
  dataset_size: 91841599
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
license: apache-2.0
language:
  - ar
tags:
  - nq
  - arabic
  - natural-questions
pretty_name: Arabic Natural Questions
size_categories:
  - 10K<n<100K

Arabic Natural Questions Dataset

This dataset is an Arabic-translated version of the original Natural Questions dataset by sentence-transformers. It contains question-answer pairs and is suitable for use with Sentence Transformers and other models requiring semantic understanding of Arabic question-answer pairs.

The dataset is designed for training and evaluating embedding models on question-answer retrieval tasks in Arabic.


📂 Dataset Subsets

pair Subset

  • Columns: "question", "answer"
  • Types: str, str

📌 Example

{
  'query': 'ما هي الوحدة الدولية لقياس شدة المجال الكهربائي؟',
  'answer': 'المجال الكهربائي هو حقل يحيط بالشحنات الكهربائية، ويُمثل القوة التي تؤثر بها شحنة على شحنة أخرى. يُعبر عن شدة المجال الكهربائي في النظام الدولي بوحدتين: نيوتن لكل كولوم (N/C)، أو فولت لكل متر (V/m). يتم إنشاء المجالات الكهربائية بواسطة الشحنات الكهربائية أو التغيرات في المجالات المغناطيسية. للمجالات الكهربائية دور مهم في الفيزياء وتُستخدم عملياً في التكنولوجيا الكهربائية. على المستوى الذري، يُعد المجال الكهربائي مسؤولاً عن التجاذب بين نواة الذرة والإلكترونات، وهو ما يساهم في ترابط الذرات وتكوين الروابط الكيميائية.',
}

Source

This dataset is based on the Natural Questions corpus and translated into Arabic for educational and research use.