Update README.md
Browse files
README.md
CHANGED
|
@@ -27,3 +27,72 @@ configs:
|
|
| 27 |
- split: train
|
| 28 |
path: data/train-*
|
| 29 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 27 |
- split: train
|
| 28 |
path: data/train-*
|
| 29 |
---
|
| 30 |
+
|
| 31 |
+
# WikiCitra-ID Dataset
|
| 32 |
+
|
| 33 |
+
WikiCitra-ID is a large-scale image-text dataset collected from Indonesian Wikipedia. The dataset consists of images and their corresponding captions extracted from the latest HTML dump of Indonesian Wikipedia articles.
|
| 34 |
+
|
| 35 |
+
## Dataset Description
|
| 36 |
+
|
| 37 |
+
WikiCitra-ID contains paired image-text data sourced from Indonesian Wikipedia articles. This dataset is derived from the image portion of [WikipediaMM-Indo](https://huggingface.co/datasets/acul3/WikipediaMM-Indo) and has been processed to create clean image-text pairs.
|
| 38 |
+
|
| 39 |
+
### Statistics
|
| 40 |
+
|
| 41 |
+
- Total number of images: 352,150
|
| 42 |
+
- Number of unique images with non-empty captions: 325,971
|
| 43 |
+
|
| 44 |
+
### Data Format
|
| 45 |
+
|
| 46 |
+
Each entry in the dataset contains:
|
| 47 |
+
- Image: Visual content from Wikipedia articles
|
| 48 |
+
- Text: Corresponding Indonesian caption or description
|
| 49 |
+
- Metadata: Additional information for verification
|
| 50 |
+
|
| 51 |
+
## Limitations and Considerations
|
| 52 |
+
|
| 53 |
+
Please be aware of the following limitations when using this dataset:
|
| 54 |
+
|
| 55 |
+
1. **Icon and Symbol Presence**
|
| 56 |
+
- The dataset may contain non-content images such as:
|
| 57 |
+
- Directional arrows
|
| 58 |
+
- National symbols
|
| 59 |
+
- UI elements
|
| 60 |
+
- Other Wikipedia icons
|
| 61 |
+
|
| 62 |
+
2. **Caption Quality**
|
| 63 |
+
- Some image-caption pairs may show inconsistencies due to crawling limitations
|
| 64 |
+
- Users can verify the alignment using provided metadata
|
| 65 |
+
- We recommend implementing additional filtering based on your specific use case
|
| 66 |
+
|
| 67 |
+
## Usage
|
| 68 |
+
|
| 69 |
+
This dataset is suitable for:
|
| 70 |
+
- Multimodal machine learning research
|
| 71 |
+
- Indonesian language vision-language models
|
| 72 |
+
- Cross-lingual image understanding
|
| 73 |
+
- Wikipedia-based knowledge extraction
|
| 74 |
+
|
| 75 |
+
## Citation
|
| 76 |
+
|
| 77 |
+
If you use this dataset in your research, please cite:
|
| 78 |
+
|
| 79 |
+
```bibtex
|
| 80 |
+
@misc{wikicitra2024,
|
| 81 |
+
title={WikiCitra-ID: A Large-scale Indonesian Wikipedia Image-Text Dataset},
|
| 82 |
+
year={2024},
|
| 83 |
+
publisher={Hugging Face}
|
| 84 |
+
}
|
| 85 |
+
```
|
| 86 |
+
|
| 87 |
+
## License
|
| 88 |
+
|
| 89 |
+
This dataset follows the same licensing as Wikipedia content. Images and text are subject to their original licenses on Wikipedia.
|
| 90 |
+
|
| 91 |
+
## Acknowledgments
|
| 92 |
+
|
| 93 |
+
- Original data sourced from Indonesian Wikipedia
|
| 94 |
+
- Built upon the WikipediaMM-Indo dataset
|
| 95 |
+
|
| 96 |
+
## Contact
|
| 97 |
+
|
| 98 |
+
For questions and feedback about the dataset, please open an issue in this repository.
|