acul3/brainly_vqa

Brainly VQA Dataset

A dataset containing 211,014 question-answer pairs scraped from Brainly.co.id, enhanced with GPT-4 generated answers for visual question answering tasks.

Dataset Description

This dataset is created by scraping educational content from Brainly.co.id and processing it through GPT-4 for enhanced answers. It contains visual questions and answers primarily in Indonesian language.

Statistics

• Total rows: 211,000
• Source: Brainly.co.id

Data Format

The dataset contains the following columns:

• __key__: Unique identifier (string)
• __url__: Original file path/location (its not valid due mistake when creating it)
• jpg: Image data (width ranges from 833-1.6k pixels)
• json: Dictionary containing:
  • conversation: Array of QA pairs
  • image_url: Source image URL
  • origin_answer: Original answer from Brainly
  • origin_url: Source URL from Brainly

Usage

This dataset is designed for:

• Visual Question Answering (VQA) tasks
• Educational AI research
• Indonesian language learning models
• Computer vision and natural language processing

Data Processing

The data collection and processing pipeline:

1. Content scraped from Brainly.co.id
2. Images extracted and stored
3. Questions and answers processed through GPT-4
4. Results formatted into structured JSON

License

This dataset is released under CC BY-NC-SA 4.0 (Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License).

Important Notes:

• The original content belongs to Brainly.co.id and its users
• This dataset is for research and non-commercial use only
• Any derivatives must be shared under the same license terms
• Attribution must be given to:
  • Brainly.co.id (original content)
  • Dataset creator (compilation and processing)

Disclaimer

This dataset is compiled from publicly available content on Brainly.co.id. The creators of this dataset do not claim ownership of the original content. If you are a content owner and wish to have your content removed, please contact the dataset maintainers.

Citation

If you use this dataset in your research, please cite:

@dataset{brainly_vqa_dataset,
  title={Brainly VQA Dataset},
  author={[Your Name]},
  year={2024},
  source={Brainly.co.id},
  size={211000},
  language={Indonesian}
}

Contact

[email protected]