license: mit
ArteFact Markdown Dataset
A comprehensive collection of scholarly art historical texts and associated images, organized by individual artworks and their corresponding scholarly documentation.
�� Dataset Overview
This dataset contains 7,200 individual art works, each with their associated scholarly markdown documentation and visual materials. The dataset serves as the primary source material for the ArteFact AI research platform, which bridges visual art and textual scholarship.
🏗️ Data Structure
artefact-markdown/
├── works/
│ ├── W1009740230/
│ │ ├── content.md # Scholarly text for this work
│ │ └── images/ # Associated visual materials
│ │ ├── image-001.jpg
│ │ ├── image-002.jpg
│ │ └── ...
│ ├── W2158127127/
│ └── ... (7,200 total work directories)
Dataset Statistics
- Total Works: 7,200
- Total Files: 239,996
- Total Size: ~19.1 GB
File Distribution by Work Size
| Files per Work | Number of Works | Percentage |
|---|---|---|
| 1 file | 1,380 | 19.2% |
| 2-10 files | 2,666 | 37.0% |
| 11-100 files | 2,800 | 38.9% |
| 100+ files | 354 | 4.9% |
| 1,000+ files | 7 | 0.1% |
| 10,000+ files | 3 | 0.04% |
Content Types
Scholarly Documentation
- Art historical analysis and descriptions
- Biographical information about artists and creators
- Historical context and provenance details
- Technical specifications and material information
- Critical interpretations and scholarly commentary
Visual Materials
- High-resolution artwork images
- Detail shots and close-ups
- Comparative images and contextual materials
- Documentation photos and archival materials
🔍 Work Identification
Each work is identified by a unique W prefix followed by a numerical identifier (e.g., W1009740230). These identifiers correspond to the works referenced in the companion artefact-json dataset.
🚀 Usage
This dataset is designed to work in conjunction with:
samwaugh/artefact-embeddings- Vector embeddings of the textual contentsamwaugh/artefact-json- Structured metadata and sentence-level annotations
Example Use Cases
- Art historical research and analysis
- AI training for art understanding models
- Scholarly text mining and analysis
- Cross-modal learning between text and images
- Digital humanities research projects
📖 Citation
If you use this dataset in your research, please cite:
@dataset{artefact_markdown_2024,
title={ArteFact Markdown Dataset},
author={Waugh, Samuel},
year={2024},
publisher={Hugging Face},
url={https://huggingface.co/datasets/samwaugh/artefact-markdown}
}
🔗 Related Resources
- ArteFact Platform: samwaugh/artefact
- Embeddings Dataset: samwaugh/artefact-embeddings
- Metadata Dataset: samwaugh/artefact-json
📄 License
This dataset is released under the MIT License. See the LICENSE file for details.
Contributing
This dataset is part of the larger ArteFact project. For questions, issues, or contributions, please refer to the main ArteFact repository.
Part of the ArteFact AI Research Platform - Bridging Visual Art and Textual Scholarship